cy0307 commited on 8 days ago

Commit

094eb82

verified ·

1 Parent(s): 557a248

Add files using upload-large-folder tool

Browse files

Files changed (27) hide show

data/artifact_index.json +29 -29
data/episode128_task_model_radar.json +52 -52
data/mirror_parity.json +493 -273
data/public_surface_qa.json +7 -7
data/quality_gates.json +1 -1
data/single_episode_task_model_radar.json +1 -1
data/source_alignment_audit.json +1 -1
data/task_method_20_gap_audit.json +16 -46
data/task_method_20_result_matrix.json +36 -36
data/task_surface_integrity.json +1 -1
data/website_integrity.json +9 -9
results/omni_finetune/model_output_probe_readiness/RUN_REPORT.md +4 -4
results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json +9 -14
results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json +187 -0
results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv +152 -0
results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv +0 -0
results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json +1 -1
results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json +1 -1
results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json +28 -0
results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv +449 -0
results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json +1 -1
results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json +1 -1
results/omni_finetune/model_output_task_probes_20260616/summary.json +17 -5
results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json +1 -1
results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json +1 -1
scripts/build_unified_task_model_radar.py +16 -2
scripts/omni/score_existing_model_output_task_probes.py +296 -6

data/artifact_index.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "title": "Ropedia Xperience-10M Task Suite Artifact Index",
-  "generated_at_utc": "2026-06-18T22:26:00+00:00",
   "status": "pass",
   "artifact_count": 214,
   "missing": [],
@@ -610,7 +610,7 @@
       "shows": "Machine-readable source-alignment pass/fail check for repo, website, and HF surfaces.",
       "exists": true,
       "bytes": 4432,
-      "sha256": "e767067c3568632f9ae600cdf123a2dae41a1cf703b7554fde7cd9b1bfee425f"
     },
     {
       "id": "source_alignment_validator",
@@ -730,8 +730,8 @@
       "surface": "website_hf",
       "shows": "Stores normalized 20-axis radar values, raw task metrics, Qwen3/Cosmos overlay mappings, branch-card caveats, and explicit scoreless status records.",
       "exists": true,
-      "bytes": 229244,
-      "sha256": "6e9de603b3f2ca60ee099135e9df559781f7481a87f6cdc7ec8873b798e14990"
     },
     {
       "id": "single_episode_task_model_radar_json",
@@ -742,7 +742,7 @@
       "shows": "Machine-readable split radar for the one-episode Minimal and Neural MLP baselines, both scored on all 20 task contracts.",
       "exists": true,
       "bytes": 51064,
-      "sha256": "6a78822649c3318371bd1943bc4aceb6216052d832b682a98d4f887dab9c2756"
     },
     {
       "id": "episode128_task_model_radar_json",
@@ -752,8 +752,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable split radar for selected 128-episode metadata/raw baselines and verified Qwen3/Cosmos branches, preserving explicit scoreless cells.",
       "exists": true,
-      "bytes": 185391,
-      "sha256": "53fbfad3e38a6accae25bb345f4682e39ebab69948df5a3423096eb7ed68b5d7"
     },
     {
       "id": "task_method_20_result_matrix_json",
@@ -763,8 +763,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable 9-method by 20-task matrix where every method has 20 records and scoreless cells carry unsupported/not-evaluated reasons.",
       "exists": true,
-      "bytes": 128842,
-      "sha256": "a361c2528ccbbd6b1b53ede78e6d39353aa98630bcfed7b8014b77b05c4f226b"
     },
     {
       "id": "task_method_20_result_matrix",
@@ -774,8 +774,8 @@
       "surface": "repo_hf",
       "shows": "Reader-facing table that separates 20 records per method from numeric scored axes, documented raw128 proxy scores, unsupported metadata targets, and model targets not evaluated in verified packages.",
       "exists": true,
-      "bytes": 3892,
-      "sha256": "590d9362624d64b81b1b19638c46faa7969730d40415e4825abd2ff09d76340a"
     },
     {
       "id": "task_method_20_gap_audit_json",
@@ -785,8 +785,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable 180-record gap ledger with numeric scores, scoreless cells, explicit status reasons, and next evidence needed before new scores can be published.",
       "exists": true,
-      "bytes": 29788,
-      "sha256": "143e03828559e706ab083b3824731d652bcb5a915cfb441caa4a3e6447a982f2"
     },
     {
       "id": "task_method_20_gap_audit",
@@ -796,8 +796,8 @@
       "surface": "repo_hf",
       "shows": "Reader-facing ledger that lists every scoreless method-task cell and the concrete target or model-output evidence required before it can become numeric.",
       "exists": true,
-      "bytes": 8787,
-      "sha256": "8fd8555f75cd96c31346618ab622e24655027cea5c52e8a09a7a53793ef7b72b"
     },
     {
       "id": "unified_task_model_radar_chart",
@@ -807,8 +807,8 @@
       "surface": "website_hf",
       "shows": "Compares minimal and neural MLP baselines across all 20 tasks, with Qwen3/Cosmos task-aligned model overlays.",
       "exists": true,
-      "bytes": 54717,
-      "sha256": "8fb6de7004e8463c28d6ada669c475d19986a1a3689778c691717cff39a7531a"
     },
     {
       "id": "single_episode_task_model_radar_chart",
@@ -829,8 +829,8 @@
       "surface": "website_hf",
       "shows": "Separates the selected 128-episode methods: raw-feature simple/NN as complete 20/20 scored polygons and metadata/Qwen/Cosmos as task-aligned overlays.",
       "exists": true,
-      "bytes": 48704,
-      "sha256": "45f8a79dc764ddbf0182ab7994fbfa5002b52b075951ab42831ecbfb3fd0405c"
     },
     {
       "id": "unified_task_model_radar_builder",
@@ -840,8 +840,8 @@
       "surface": "repo_hf",
       "shows": "Regenerates the direction-aware radar chart and machine-readable metric overlay JSON.",
       "exists": true,
-      "bytes": 56295,
-      "sha256": "237ef6a08dd81b9fcb0256f6f6b54394f74c14daea0cbba8f1726fdbbb7feff0"
     },
     {
       "id": "task_method_20_gap_audit_builder",
@@ -873,8 +873,8 @@
       "surface": "repo_hf",
       "shows": "Checks whether Qwen3/Cosmos branches have train, validation, and test prediction files before extending model overlays to all 20 task contracts.",
       "exists": true,
-      "bytes": 5144,
-      "sha256": "86835a291c659d335ab5b51d09ea3f36710c14d23466a72d9c2a23137dc9ae8a"
     },
     {
       "id": "model_output_probe_script",
@@ -895,8 +895,8 @@
       "surface": "repo_hf",
       "shows": "Scores task-specific Qwen3/Cosmos overlays only where verified held-out prediction JSON or compact target maps already contain the required targets.",
       "exists": true,
-      "bytes": 5278,
-      "sha256": "4a27f39807e8ac13771e2109584a8b0265961bcea88445058cbfe87265457d7f"
     },
     {
       "id": "existing_model_output_task_probe_script",
@@ -906,8 +906,8 @@
       "surface": "repo_hf",
       "shows": "Derives task-specific scores from committed verified model outputs without running new inference or backfilling absent targets.",
       "exists": true,
-      "bytes": 57470,
-      "sha256": "9a0708885c1a22addc0bec199488021e277c7a7b7e23d6a37344fbbb3e5f1b96"
     },
     {
       "id": "a100_128_metadata_task_baselines",
@@ -1116,7 +1116,7 @@
       "shows": "Machine-readable release-check summary for validators, mirrors, and public project surfaces.",
       "exists": true,
       "bytes": 8100,
-      "sha256": "12cec90cfb4bd783f397852f04ffba5b9d0ae9316667bdee4719cbd6f2d3579a"
     },
     {
       "id": "public_surface_qa",
@@ -1321,7 +1321,7 @@
       "volatile": true,
       "shows": "Confirms prepared GitHub/HF Space/artifact/model mirrors share the same critical data, figure, website HTML, and validator files.",
       "exists": true,
-      "bytes": 938745,
       "hash_policy": "existence_and_size_only"
     },
     {

 {
   "title": "Ropedia Xperience-10M Task Suite Artifact Index",
+  "generated_at_utc": "2026-06-18T22:57:14+00:00",
   "status": "pass",
   "artifact_count": 214,
   "missing": [],
       "shows": "Machine-readable source-alignment pass/fail check for repo, website, and HF surfaces.",
       "exists": true,
       "bytes": 4432,
+      "sha256": "1df30908392b455cc7263154126d55709cf97c60ff4213f0eee9ce17d92fb234"
     },
     {
       "id": "source_alignment_validator",
       "surface": "website_hf",
       "shows": "Stores normalized 20-axis radar values, raw task metrics, Qwen3/Cosmos overlay mappings, branch-card caveats, and explicit scoreless status records.",
       "exists": true,
+      "bytes": 229168,
+      "sha256": "64c7c10f3ce663178f2eea4740bb45de0f94b1246448f14f5521d293be916e77"
     },
     {
       "id": "single_episode_task_model_radar_json",
       "shows": "Machine-readable split radar for the one-episode Minimal and Neural MLP baselines, both scored on all 20 task contracts.",
       "exists": true,
       "bytes": 51064,
+      "sha256": "9d42b045d833e50a2bfa8905203150af1f28ea4736d22edb0fe27e39eedf4574"
     },
     {
       "id": "episode128_task_model_radar_json",
       "surface": "website_hf",
       "shows": "Machine-readable split radar for selected 128-episode metadata/raw baselines and verified Qwen3/Cosmos branches, preserving explicit scoreless cells.",
       "exists": true,
+      "bytes": 185314,
+      "sha256": "5c61c8f9a4a9a7b6bdc65e39ef5fda6117df48412afe260c0be4da501bd7e8db"
     },
     {
       "id": "task_method_20_result_matrix_json",
       "surface": "website_hf",
       "shows": "Machine-readable 9-method by 20-task matrix where every method has 20 records and scoreless cells carry unsupported/not-evaluated reasons.",
       "exists": true,
+      "bytes": 128805,
+      "sha256": "d5572768c6fb671eadf87870d20c8645dfae384f862aa63e929b2db50a5cc51a"
     },
     {
       "id": "task_method_20_result_matrix",
       "surface": "repo_hf",
       "shows": "Reader-facing table that separates 20 records per method from numeric scored axes, documented raw128 proxy scores, unsupported metadata targets, and model targets not evaluated in verified packages.",
       "exists": true,
+      "bytes": 3876,
+      "sha256": "a9ce23a452ca73b6d08219ffaac1782d5baad11f9e3662138b62cb4cd5ff206f"
     },
     {
       "id": "task_method_20_gap_audit_json",
       "surface": "website_hf",
       "shows": "Machine-readable 180-record gap ledger with numeric scores, scoreless cells, explicit status reasons, and next evidence needed before new scores can be published.",
       "exists": true,
+      "bytes": 28259,
+      "sha256": "860dea3d3ce2f2c2146c97918a5deca05ae315238b584c10e53f74a9d2c26dcd"
     },
     {
       "id": "task_method_20_gap_audit",
       "surface": "repo_hf",
       "shows": "Reader-facing ledger that lists every scoreless method-task cell and the concrete target or model-output evidence required before it can become numeric.",
       "exists": true,
+      "bytes": 8428,
+      "sha256": "b33c7f91565a2603b138eb4ec1c6df32d823225b16563eba279f416b0fe2a964"
     },
     {
       "id": "unified_task_model_radar_chart",
       "surface": "website_hf",
       "shows": "Compares minimal and neural MLP baselines across all 20 tasks, with Qwen3/Cosmos task-aligned model overlays.",
       "exists": true,
+      "bytes": 54936,
+      "sha256": "f29db4099d60c093ebd092bee37a2fc1338bbff17d7e52b8a1b27ae6fc06727a"
     },
     {
       "id": "single_episode_task_model_radar_chart",
       "surface": "website_hf",
       "shows": "Separates the selected 128-episode methods: raw-feature simple/NN as complete 20/20 scored polygons and metadata/Qwen/Cosmos as task-aligned overlays.",
       "exists": true,
+      "bytes": 48923,
+      "sha256": "75c389122a413548bdc26fd05cbbad2a85015ebd222ba1a23571f3ff4f0eddcf"
     },
     {
       "id": "unified_task_model_radar_builder",
       "surface": "repo_hf",
       "shows": "Regenerates the direction-aware radar chart and machine-readable metric overlay JSON.",
       "exists": true,
+      "bytes": 57104,
+      "sha256": "805cfce89529fec73edd3e436c1a76afae46cebe31ba5e441b6dfb5cc142c935"
     },
     {
       "id": "task_method_20_gap_audit_builder",
       "surface": "repo_hf",
       "shows": "Checks whether Qwen3/Cosmos branches have train, validation, and test prediction files before extending model overlays to all 20 task contracts.",
       "exists": true,
+      "bytes": 4977,
+      "sha256": "b9caf6d32bb801d845b57b4450b92c10a4c3f9d14890f6a209e5a52aede92568"
     },
     {
       "id": "model_output_probe_script",
       "surface": "repo_hf",
       "shows": "Scores task-specific Qwen3/Cosmos overlays only where verified held-out prediction JSON or compact target maps already contain the required targets.",
       "exists": true,
+      "bytes": 5951,
+      "sha256": "910477d2fba648605dda128d0ecd2a2c13cfa460573e350dc850014ac91c6c2b"
     },
     {
       "id": "existing_model_output_task_probe_script",
       "surface": "repo_hf",
       "shows": "Derives task-specific scores from committed verified model outputs without running new inference or backfilling absent targets.",
       "exists": true,
+      "bytes": 69411,
+      "sha256": "a78889e5225405fcbb0156aa8c63cbbe79022a12283d0b398c5f2bb015175fe7"
     },
     {
       "id": "a100_128_metadata_task_baselines",
       "shows": "Machine-readable release-check summary for validators, mirrors, and public project surfaces.",
       "exists": true,
       "bytes": 8100,
+      "sha256": "b88b46bd9b16d4afaf4836d0acff208b46559a493cfcffa881f1528fa282292c"
     },
     {
       "id": "public_surface_qa",
       "volatile": true,
       "shows": "Confirms prepared GitHub/HF Space/artifact/model mirrors share the same critical data, figure, website HTML, and validator files.",
       "exists": true,
+      "bytes": 1015184,
       "hash_policy": "existence_and_size_only"
     },
     {

data/episode128_task_model_radar.json CHANGED Viewed

@@ -1,12 +1,12 @@
 {
   "title": "128-Episode 20-Task Radar",
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:25:59+00:00",
   "description": "Selected 128-episode metadata/raw baselines plus verified Qwen3/Cosmos branches. Every method has 20 records; numeric scores appear only where the public artifact produced that task target.",
   "task_count": 20,
   "method_count": 7,
   "method_task_record_count": 140,
-  "scored_method_task_count": 111,
   "normalization_policy": {
     "higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
     "lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
@@ -147,20 +147,20 @@
       "kind": "partial_128_episode_foundation_model_overlay",
       "scope": "128 selected episodes, held-out test",
       "stroke_dasharray": "4 7",
-      "method_detail": "Verified Cosmos3-Super base-weight Reasoner JSON-task evaluation, plus task 16 and a derived task-20 action-boundary timing probe scored from existing verified JSON.",
       "plotted_as": "colored point overlay",
       "result_record_count": 20,
-      "scored_task_count": 9,
-      "covered_task_count": 9,
       "proxy_scored_task_count": 0,
-      "scoreless_task_count": 11,
       "unsupported_task_count": 0,
-      "not_evaluated_task_count": 11,
       "status_counts": {
-        "not_evaluated_in_verified_package": 11,
-        "scored": 9
       },
-      "coverage_fraction": 0.45,
       "result_record_fraction": 1.0
     },
     {
@@ -171,20 +171,20 @@
       "kind": "partial_128_episode_world_model_overlay",
       "scope": "128 selected episodes, held-out test",
       "stroke_dasharray": "2 7",
-      "method_detail": "Verified Cosmos3-Nano future-window compatibility metrics, plus tasks 10/13/14/17 and a derived task-20 boundary timing probe scored from existing held-out future-window artifacts.",
       "plotted_as": "colored point overlay",
       "result_record_count": 20,
-      "scored_task_count": 10,
-      "covered_task_count": 10,
       "proxy_scored_task_count": 0,
-      "scoreless_task_count": 10,
       "unsupported_task_count": 0,
-      "not_evaluated_task_count": 10,
       "status_counts": {
-        "not_evaluated_in_verified_package": 10,
-        "scored": 10
       },
-      "coverage_fraction": 0.5,
       "result_record_fraction": 1.0
     }
   ],
@@ -894,15 +894,15 @@
           "status_label": "scored"
         },
         "cosmos3_super_reasoner": {
-          "raw": null,
-          "metric_key": "mrr",
-          "source": null,
           "scope": "multi_episode_128_partial_model_overlay",
-          "status": "not_evaluated_in_verified_package",
-          "reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
-          "normalized_score": null,
-          "raw_text": "n/a",
-          "status_label": "not evaluated"
         },
         "cosmos3_nano_future_window": {
           "raw": null,
@@ -1633,15 +1633,15 @@
           "status_label": "scored"
         },
         "cosmos3_nano_future_window": {
-          "raw": null,
-          "metric_key": "macro_f1",
-          "source": null,
           "scope": "multi_episode_128_partial_model_overlay",
-          "status": "not_evaluated_in_verified_package",
-          "reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
-          "normalized_score": null,
-          "raw_text": "n/a",
-          "status_label": "not evaluated"
         }
       }
     },
@@ -2989,17 +2989,17 @@
       "task_label": "Language Grounding",
       "series_id": "cosmos3_super_reasoner",
       "method": "Cosmos3-Super Reasoner",
-      "status": "not_evaluated_in_verified_package",
-      "status_label": "not evaluated",
-      "scored": false,
       "proxy_scored": false,
-      "raw": null,
-      "raw_text": "n/a",
-      "normalized_score": null,
-      "metric_key": "mrr",
-      "source": null,
       "scope": "multi_episode_128_partial_model_overlay",
-      "reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score"
     },
     {
       "task_number": 8,
@@ -4015,17 +4015,17 @@
       "task_label": "Action-Object Relation Prediction",
       "series_id": "cosmos3_nano_future_window",
       "method": "Cosmos3-Nano Future Window",
-      "status": "not_evaluated_in_verified_package",
-      "status_label": "not evaluated",
-      "scored": false,
       "proxy_scored": false,
-      "raw": null,
-      "raw_text": "n/a",
-      "normalized_score": null,
-      "metric_key": "macro_f1",
-      "source": null,
       "scope": "multi_episode_128_partial_model_overlay",
-      "reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score"
     },
     {
       "task_number": 17,

 {
   "title": "128-Episode 20-Task Radar",
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:55:01+00:00",
   "description": "Selected 128-episode metadata/raw baselines plus verified Qwen3/Cosmos branches. Every method has 20 records; numeric scores appear only where the public artifact produced that task target.",
   "task_count": 20,
   "method_count": 7,
   "method_task_record_count": 140,
+  "scored_method_task_count": 113,
   "normalization_policy": {
     "higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
     "lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
       "kind": "partial_128_episode_foundation_model_overlay",
       "scope": "128 selected episodes, held-out test",
       "stroke_dasharray": "4 7",
+      "method_detail": "Verified Cosmos3-Super base-weight Reasoner JSON-task evaluation, plus task 8/16 and a derived task-20 action-boundary timing probe scored from existing verified JSON.",
       "plotted_as": "colored point overlay",
       "result_record_count": 20,
+      "scored_task_count": 10,
+      "covered_task_count": 10,
       "proxy_scored_task_count": 0,
+      "scoreless_task_count": 10,
       "unsupported_task_count": 0,
+      "not_evaluated_task_count": 10,
       "status_counts": {
+        "not_evaluated_in_verified_package": 10,
+        "scored": 10
       },
+      "coverage_fraction": 0.5,
       "result_record_fraction": 1.0
     },
     {
       "kind": "partial_128_episode_world_model_overlay",
       "scope": "128 selected episodes, held-out test",
       "stroke_dasharray": "2 7",
+      "method_detail": "Verified Cosmos3-Nano future-window compatibility metrics, plus tasks 10/13/14/16/17 and a derived task-20 boundary timing probe scored from existing held-out future-window artifacts.",
       "plotted_as": "colored point overlay",
       "result_record_count": 20,
+      "scored_task_count": 11,
+      "covered_task_count": 11,
       "proxy_scored_task_count": 0,
+      "scoreless_task_count": 9,
       "unsupported_task_count": 0,
+      "not_evaluated_task_count": 9,
       "status_counts": {
+        "not_evaluated_in_verified_package": 9,
+        "scored": 11
       },
+      "coverage_fraction": 0.55,
       "result_record_fraction": 1.0
     }
   ],
           "status_label": "scored"
         },
         "cosmos3_super_reasoner": {
+          "raw": 0.30639899644580487,
+          "metric_key": "caption_grounding_iou",
+          "source": "results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
           "scope": "multi_episode_128_partial_model_overlay",
+          "status": "scored",
+          "reason": null,
+          "normalized_score": 0.30639899644580487,
+          "raw_text": "0.3064",
+          "status_label": "scored"
         },
         "cosmos3_nano_future_window": {
           "raw": null,
           "status_label": "scored"
         },
         "cosmos3_nano_future_window": {
+          "raw": 0.002794157670325683,
+          "metric_key": "action_object_relation_macro_f1",
+          "source": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
           "scope": "multi_episode_128_partial_model_overlay",
+          "status": "scored",
+          "reason": null,
+          "normalized_score": 0.002794157670325683,
+          "raw_text": "0.0028",
+          "status_label": "scored"
         }
       }
     },
       "task_label": "Language Grounding",
       "series_id": "cosmos3_super_reasoner",
       "method": "Cosmos3-Super Reasoner",
+      "status": "scored",
+      "status_label": "scored",
+      "scored": true,
       "proxy_scored": false,
+      "raw": 0.30639899644580487,
+      "raw_text": "0.3064",
+      "normalized_score": 0.30639899644580487,
+      "metric_key": "caption_grounding_iou",
+      "source": "results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
       "scope": "multi_episode_128_partial_model_overlay",
+      "reason": null
     },
     {
       "task_number": 8,
       "task_label": "Action-Object Relation Prediction",
       "series_id": "cosmos3_nano_future_window",
       "method": "Cosmos3-Nano Future Window",
+      "status": "scored",
+      "status_label": "scored",
+      "scored": true,
       "proxy_scored": false,
+      "raw": 0.002794157670325683,
+      "raw_text": "0.0028",
+      "normalized_score": 0.002794157670325683,
+      "metric_key": "action_object_relation_macro_f1",
+      "source": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
       "scope": "multi_episode_128_partial_model_overlay",
+      "reason": null
     },
     {
       "task_number": 17,

data/mirror_parity.json CHANGED Viewed

@@ -1,9 +1,9 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:36:57+00:00",
   "hf_root": "hf_publish",
   "summary": {
-    "group_count": 855,
     "failure_count": 0,
     "failures_by_surface": {}
   },
@@ -138,45 +138,45 @@
       "local": {
         "path": "repo:docs/data/artifact_index.json",
         "exists": true,
-        "bytes": 116643,
-        "sha256": "52e9ef2ca8b7bdcb1ab53f5bbed55a1a1e846eb7c4dbf13eaf25b210f581e2bd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/artifact_index.json",
           "exists": true,
-          "bytes": 116643,
-          "sha256": "52e9ef2ca8b7bdcb1ab53f5bbed55a1a1e846eb7c4dbf13eaf25b210f581e2bd"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/artifact_index.json",
           "exists": true,
-          "bytes": 116643,
-          "sha256": "52e9ef2ca8b7bdcb1ab53f5bbed55a1a1e846eb7c4dbf13eaf25b210f581e2bd"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/artifact_index.json",
           "exists": true,
-          "bytes": 116643,
-          "sha256": "52e9ef2ca8b7bdcb1ab53f5bbed55a1a1e846eb7c4dbf13eaf25b210f581e2bd"
         },
         "hf_model_data": {
           "path": "hf_model:data/artifact_index.json",
           "exists": true,
-          "bytes": 116643,
-          "sha256": "52e9ef2ca8b7bdcb1ab53f5bbed55a1a1e846eb7c4dbf13eaf25b210f581e2bd"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/artifact_index.json",
           "exists": true,
-          "bytes": 116643,
-          "sha256": "52e9ef2ca8b7bdcb1ab53f5bbed55a1a1e846eb7c4dbf13eaf25b210f581e2bd"
         },
         "hf_model": {
           "path": "hf_model:metrics/artifact_index.json",
           "exists": true,
-          "bytes": 116643,
-          "sha256": "52e9ef2ca8b7bdcb1ab53f5bbed55a1a1e846eb7c4dbf13eaf25b210f581e2bd"
         }
       },
       "failures": []
@@ -922,45 +922,45 @@
       "local": {
         "path": "repo:docs/data/publication_audit.json",
         "exists": true,
-        "bytes": 9613,
-        "sha256": "1f9344871b3783f39e6ed709d892cf0e60c2b2d29fcd635f711672625a3404b1"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/publication_audit.json",
           "exists": true,
-          "bytes": 9613,
-          "sha256": "1f9344871b3783f39e6ed709d892cf0e60c2b2d29fcd635f711672625a3404b1"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/publication_audit.json",
           "exists": true,
-          "bytes": 9613,
-          "sha256": "1f9344871b3783f39e6ed709d892cf0e60c2b2d29fcd635f711672625a3404b1"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/publication_audit.json",
           "exists": true,
-          "bytes": 9613,
-          "sha256": "1f9344871b3783f39e6ed709d892cf0e60c2b2d29fcd635f711672625a3404b1"
         },
         "hf_model_data": {
           "path": "hf_model:data/publication_audit.json",
           "exists": true,
-          "bytes": 9613,
-          "sha256": "1f9344871b3783f39e6ed709d892cf0e60c2b2d29fcd635f711672625a3404b1"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/publication_audit.json",
           "exists": true,
-          "bytes": 9613,
-          "sha256": "1f9344871b3783f39e6ed709d892cf0e60c2b2d29fcd635f711672625a3404b1"
         },
         "hf_model": {
           "path": "hf_model:metrics/publication_audit.json",
           "exists": true,
-          "bytes": 9613,
-          "sha256": "1f9344871b3783f39e6ed709d892cf0e60c2b2d29fcd635f711672625a3404b1"
         }
       },
       "failures": []
@@ -972,44 +972,44 @@
         "path": "repo:docs/data/public_surface_qa.json",
         "exists": true,
         "bytes": 7126,
-        "sha256": "1f57b699b7ad919698e69b8cdef576cd525878468c6d8e86621431eb15402c5e"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
-          "sha256": "1f57b699b7ad919698e69b8cdef576cd525878468c6d8e86621431eb15402c5e"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
-          "sha256": "1f57b699b7ad919698e69b8cdef576cd525878468c6d8e86621431eb15402c5e"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
-          "sha256": "1f57b699b7ad919698e69b8cdef576cd525878468c6d8e86621431eb15402c5e"
         },
         "hf_model_data": {
           "path": "hf_model:data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
-          "sha256": "1f57b699b7ad919698e69b8cdef576cd525878468c6d8e86621431eb15402c5e"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
-          "sha256": "1f57b699b7ad919698e69b8cdef576cd525878468c6d8e86621431eb15402c5e"
         },
         "hf_model": {
           "path": "hf_model:metrics/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
-          "sha256": "1f57b699b7ad919698e69b8cdef576cd525878468c6d8e86621431eb15402c5e"
         }
       },
       "failures": []
@@ -1119,44 +1119,44 @@
         "path": "repo:docs/data/quality_gates.json",
         "exists": true,
         "bytes": 8100,
-        "sha256": "16a1241e753b46169ee937ed28b00dc63a41509c6d127a9068aabd831e9e40cc"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
-          "sha256": "16a1241e753b46169ee937ed28b00dc63a41509c6d127a9068aabd831e9e40cc"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
-          "sha256": "16a1241e753b46169ee937ed28b00dc63a41509c6d127a9068aabd831e9e40cc"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
-          "sha256": "16a1241e753b46169ee937ed28b00dc63a41509c6d127a9068aabd831e9e40cc"
         },
         "hf_model_data": {
           "path": "hf_model:data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
-          "sha256": "16a1241e753b46169ee937ed28b00dc63a41509c6d127a9068aabd831e9e40cc"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
-          "sha256": "16a1241e753b46169ee937ed28b00dc63a41509c6d127a9068aabd831e9e40cc"
         },
         "hf_model": {
           "path": "hf_model:metrics/quality_gates.json",
           "exists": true,
           "bytes": 8100,
-          "sha256": "16a1241e753b46169ee937ed28b00dc63a41509c6d127a9068aabd831e9e40cc"
         }
       },
       "failures": []
@@ -1560,44 +1560,44 @@
         "path": "repo:docs/data/scope_claims_audit.json",
         "exists": true,
         "bytes": 21630,
-        "sha256": "2fef6d1cee15981aec3aa4af933c767997e01500a0fd6141f61d144223e24ad3"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
-          "sha256": "2fef6d1cee15981aec3aa4af933c767997e01500a0fd6141f61d144223e24ad3"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
-          "sha256": "2fef6d1cee15981aec3aa4af933c767997e01500a0fd6141f61d144223e24ad3"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
-          "sha256": "2fef6d1cee15981aec3aa4af933c767997e01500a0fd6141f61d144223e24ad3"
         },
         "hf_model_data": {
           "path": "hf_model:data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
-          "sha256": "2fef6d1cee15981aec3aa4af933c767997e01500a0fd6141f61d144223e24ad3"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
-          "sha256": "2fef6d1cee15981aec3aa4af933c767997e01500a0fd6141f61d144223e24ad3"
         },
         "hf_model": {
           "path": "hf_model:metrics/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
-          "sha256": "2fef6d1cee15981aec3aa4af933c767997e01500a0fd6141f61d144223e24ad3"
         }
       },
       "failures": []
@@ -1658,44 +1658,44 @@
         "path": "repo:docs/data/source_alignment_audit.json",
         "exists": true,
         "bytes": 4432,
-        "sha256": "1df30908392b455cc7263154126d55709cf97c60ff4213f0eee9ce17d92fb234"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "1df30908392b455cc7263154126d55709cf97c60ff4213f0eee9ce17d92fb234"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "1df30908392b455cc7263154126d55709cf97c60ff4213f0eee9ce17d92fb234"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "1df30908392b455cc7263154126d55709cf97c60ff4213f0eee9ce17d92fb234"
         },
         "hf_model_data": {
           "path": "hf_model:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "1df30908392b455cc7263154126d55709cf97c60ff4213f0eee9ce17d92fb234"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "1df30908392b455cc7263154126d55709cf97c60ff4213f0eee9ce17d92fb234"
         },
         "hf_model": {
           "path": "hf_model:metrics/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "1df30908392b455cc7263154126d55709cf97c60ff4213f0eee9ce17d92fb234"
         }
       },
       "failures": []
@@ -1756,44 +1756,44 @@
         "path": "repo:docs/data/single_episode_task_model_radar.json",
         "exists": true,
         "bytes": 51064,
-        "sha256": "6a78822649c3318371bd1943bc4aceb6216052d832b682a98d4f887dab9c2756"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
-          "sha256": "6a78822649c3318371bd1943bc4aceb6216052d832b682a98d4f887dab9c2756"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
-          "sha256": "6a78822649c3318371bd1943bc4aceb6216052d832b682a98d4f887dab9c2756"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
-          "sha256": "6a78822649c3318371bd1943bc4aceb6216052d832b682a98d4f887dab9c2756"
         },
         "hf_model_data": {
           "path": "hf_model:data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
-          "sha256": "6a78822649c3318371bd1943bc4aceb6216052d832b682a98d4f887dab9c2756"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
-          "sha256": "6a78822649c3318371bd1943bc4aceb6216052d832b682a98d4f887dab9c2756"
         },
         "hf_model": {
           "path": "hf_model:metrics/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
-          "sha256": "6a78822649c3318371bd1943bc4aceb6216052d832b682a98d4f887dab9c2756"
         }
       },
       "failures": []
@@ -1804,45 +1804,45 @@
       "local": {
         "path": "repo:docs/data/episode128_task_model_radar.json",
         "exists": true,
-        "bytes": 185391,
-        "sha256": "53fbfad3e38a6accae25bb345f4682e39ebab69948df5a3423096eb7ed68b5d7"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/episode128_task_model_radar.json",
           "exists": true,
-          "bytes": 185391,
-          "sha256": "53fbfad3e38a6accae25bb345f4682e39ebab69948df5a3423096eb7ed68b5d7"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/episode128_task_model_radar.json",
           "exists": true,
-          "bytes": 185391,
-          "sha256": "53fbfad3e38a6accae25bb345f4682e39ebab69948df5a3423096eb7ed68b5d7"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/episode128_task_model_radar.json",
           "exists": true,
-          "bytes": 185391,
-          "sha256": "53fbfad3e38a6accae25bb345f4682e39ebab69948df5a3423096eb7ed68b5d7"
         },
         "hf_model_data": {
           "path": "hf_model:data/episode128_task_model_radar.json",
           "exists": true,
-          "bytes": 185391,
-          "sha256": "53fbfad3e38a6accae25bb345f4682e39ebab69948df5a3423096eb7ed68b5d7"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/episode128_task_model_radar.json",
           "exists": true,
-          "bytes": 185391,
-          "sha256": "53fbfad3e38a6accae25bb345f4682e39ebab69948df5a3423096eb7ed68b5d7"
         },
         "hf_model": {
           "path": "hf_model:metrics/episode128_task_model_radar.json",
           "exists": true,
-          "bytes": 185391,
-          "sha256": "53fbfad3e38a6accae25bb345f4682e39ebab69948df5a3423096eb7ed68b5d7"
         }
       },
       "failures": []
@@ -2001,44 +2001,44 @@
         "path": "repo:docs/data/task_surface_integrity.json",
         "exists": true,
         "bytes": 45779,
-        "sha256": "4c0b7c81b916d32d7b2d8453c5b3e8dbf5e19d0e4002c966b9d4a34f33e49f0d"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "4c0b7c81b916d32d7b2d8453c5b3e8dbf5e19d0e4002c966b9d4a34f33e49f0d"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "4c0b7c81b916d32d7b2d8453c5b3e8dbf5e19d0e4002c966b9d4a34f33e49f0d"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "4c0b7c81b916d32d7b2d8453c5b3e8dbf5e19d0e4002c966b9d4a34f33e49f0d"
         },
         "hf_model_data": {
           "path": "hf_model:data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "4c0b7c81b916d32d7b2d8453c5b3e8dbf5e19d0e4002c966b9d4a34f33e49f0d"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "4c0b7c81b916d32d7b2d8453c5b3e8dbf5e19d0e4002c966b9d4a34f33e49f0d"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "4c0b7c81b916d32d7b2d8453c5b3e8dbf5e19d0e4002c966b9d4a34f33e49f0d"
         }
       },
       "failures": []
@@ -2098,45 +2098,45 @@
       "local": {
         "path": "repo:docs/data/task_method_20_result_matrix.json",
         "exists": true,
-        "bytes": 128842,
-        "sha256": "a361c2528ccbbd6b1b53ede78e6d39353aa98630bcfed7b8014b77b05c4f226b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_method_20_result_matrix.json",
           "exists": true,
-          "bytes": 128842,
-          "sha256": "a361c2528ccbbd6b1b53ede78e6d39353aa98630bcfed7b8014b77b05c4f226b"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/task_method_20_result_matrix.json",
           "exists": true,
-          "bytes": 128842,
-          "sha256": "a361c2528ccbbd6b1b53ede78e6d39353aa98630bcfed7b8014b77b05c4f226b"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_method_20_result_matrix.json",
           "exists": true,
-          "bytes": 128842,
-          "sha256": "a361c2528ccbbd6b1b53ede78e6d39353aa98630bcfed7b8014b77b05c4f226b"
         },
         "hf_model_data": {
           "path": "hf_model:data/task_method_20_result_matrix.json",
           "exists": true,
-          "bytes": 128842,
-          "sha256": "a361c2528ccbbd6b1b53ede78e6d39353aa98630bcfed7b8014b77b05c4f226b"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/task_method_20_result_matrix.json",
           "exists": true,
-          "bytes": 128842,
-          "sha256": "a361c2528ccbbd6b1b53ede78e6d39353aa98630bcfed7b8014b77b05c4f226b"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_method_20_result_matrix.json",
           "exists": true,
-          "bytes": 128842,
-          "sha256": "a361c2528ccbbd6b1b53ede78e6d39353aa98630bcfed7b8014b77b05c4f226b"
         }
       },
       "failures": []
@@ -2147,45 +2147,45 @@
       "local": {
         "path": "repo:docs/data/task_method_20_gap_audit.json",
         "exists": true,
-        "bytes": 29788,
-        "sha256": "143e03828559e706ab083b3824731d652bcb5a915cfb441caa4a3e6447a982f2"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_method_20_gap_audit.json",
           "exists": true,
-          "bytes": 29788,
-          "sha256": "143e03828559e706ab083b3824731d652bcb5a915cfb441caa4a3e6447a982f2"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/task_method_20_gap_audit.json",
           "exists": true,
-          "bytes": 29788,
-          "sha256": "143e03828559e706ab083b3824731d652bcb5a915cfb441caa4a3e6447a982f2"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_method_20_gap_audit.json",
           "exists": true,
-          "bytes": 29788,
-          "sha256": "143e03828559e706ab083b3824731d652bcb5a915cfb441caa4a3e6447a982f2"
         },
         "hf_model_data": {
           "path": "hf_model:data/task_method_20_gap_audit.json",
           "exists": true,
-          "bytes": 29788,
-          "sha256": "143e03828559e706ab083b3824731d652bcb5a915cfb441caa4a3e6447a982f2"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/task_method_20_gap_audit.json",
           "exists": true,
-          "bytes": 29788,
-          "sha256": "143e03828559e706ab083b3824731d652bcb5a915cfb441caa4a3e6447a982f2"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_method_20_gap_audit.json",
           "exists": true,
-          "bytes": 29788,
-          "sha256": "143e03828559e706ab083b3824731d652bcb5a915cfb441caa4a3e6447a982f2"
         }
       },
       "failures": []
@@ -2245,45 +2245,45 @@
       "local": {
         "path": "repo:docs/data/unified_task_model_radar.json",
         "exists": true,
-        "bytes": 229244,
-        "sha256": "6e9de603b3f2ca60ee099135e9df559781f7481a87f6cdc7ec8873b798e14990"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/unified_task_model_radar.json",
           "exists": true,
-          "bytes": 229244,
-          "sha256": "6e9de603b3f2ca60ee099135e9df559781f7481a87f6cdc7ec8873b798e14990"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/unified_task_model_radar.json",
           "exists": true,
-          "bytes": 229244,
-          "sha256": "6e9de603b3f2ca60ee099135e9df559781f7481a87f6cdc7ec8873b798e14990"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/unified_task_model_radar.json",
           "exists": true,
-          "bytes": 229244,
-          "sha256": "6e9de603b3f2ca60ee099135e9df559781f7481a87f6cdc7ec8873b798e14990"
         },
         "hf_model_data": {
           "path": "hf_model:data/unified_task_model_radar.json",
           "exists": true,
-          "bytes": 229244,
-          "sha256": "6e9de603b3f2ca60ee099135e9df559781f7481a87f6cdc7ec8873b798e14990"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/unified_task_model_radar.json",
           "exists": true,
-          "bytes": 229244,
-          "sha256": "6e9de603b3f2ca60ee099135e9df559781f7481a87f6cdc7ec8873b798e14990"
         },
         "hf_model": {
           "path": "hf_model:metrics/unified_task_model_radar.json",
           "exists": true,
-          "bytes": 229244,
-          "sha256": "6e9de603b3f2ca60ee099135e9df559781f7481a87f6cdc7ec8873b798e14990"
         }
       },
       "failures": []
@@ -2294,45 +2294,45 @@
       "local": {
         "path": "repo:docs/data/website_integrity.json",
         "exists": true,
-        "bytes": 20021,
-        "sha256": "f9b0ce37c945cd95d809eb447ac950464ec6e450411ecd816618a79f83581bbd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/website_integrity.json",
           "exists": true,
-          "bytes": 20021,
-          "sha256": "f9b0ce37c945cd95d809eb447ac950464ec6e450411ecd816618a79f83581bbd"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/website_integrity.json",
           "exists": true,
-          "bytes": 20021,
-          "sha256": "f9b0ce37c945cd95d809eb447ac950464ec6e450411ecd816618a79f83581bbd"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/website_integrity.json",
           "exists": true,
-          "bytes": 20021,
-          "sha256": "f9b0ce37c945cd95d809eb447ac950464ec6e450411ecd816618a79f83581bbd"
         },
         "hf_model_data": {
           "path": "hf_model:data/website_integrity.json",
           "exists": true,
-          "bytes": 20021,
-          "sha256": "f9b0ce37c945cd95d809eb447ac950464ec6e450411ecd816618a79f83581bbd"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/website_integrity.json",
           "exists": true,
-          "bytes": 20021,
-          "sha256": "f9b0ce37c945cd95d809eb447ac950464ec6e450411ecd816618a79f83581bbd"
         },
         "hf_model": {
           "path": "hf_model:metrics/website_integrity.json",
           "exists": true,
-          "bytes": 20021,
-          "sha256": "f9b0ce37c945cd95d809eb447ac950464ec6e450411ecd816618a79f83581bbd"
         }
       },
       "failures": []
@@ -2515,33 +2515,33 @@
       "local": {
         "path": "repo:docs/assets/charts/episode128_task_model_radar.svg",
         "exists": true,
-        "bytes": 48704,
-        "sha256": "45f8a79dc764ddbf0182ab7994fbfa5002b52b075951ab42831ecbfb3fd0405c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:assets/charts/episode128_task_model_radar.svg",
           "exists": true,
-          "bytes": 48704,
-          "sha256": "45f8a79dc764ddbf0182ab7994fbfa5002b52b075951ab42831ecbfb3fd0405c"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/assets/charts/episode128_task_model_radar.svg",
           "exists": true,
-          "bytes": 48704,
-          "sha256": "45f8a79dc764ddbf0182ab7994fbfa5002b52b075951ab42831ecbfb3fd0405c"
         },
         "hf_artifacts_card": {
           "path": "hf_artifacts:assets/charts/episode128_task_model_radar.svg",
           "exists": true,
-          "bytes": 48704,
-          "sha256": "45f8a79dc764ddbf0182ab7994fbfa5002b52b075951ab42831ecbfb3fd0405c"
         },
         "hf_model": {
           "path": "hf_model:assets/charts/episode128_task_model_radar.svg",
           "exists": true,
-          "bytes": 48704,
-          "sha256": "45f8a79dc764ddbf0182ab7994fbfa5002b52b075951ab42831ecbfb3fd0405c"
         }
       },
       "failures": []
@@ -2589,33 +2589,33 @@
       "local": {
         "path": "repo:docs/assets/charts/unified_task_model_radar.svg",
         "exists": true,
-        "bytes": 54717,
-        "sha256": "8fb6de7004e8463c28d6ada669c475d19986a1a3689778c691717cff39a7531a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:assets/charts/unified_task_model_radar.svg",
           "exists": true,
-          "bytes": 54717,
-          "sha256": "8fb6de7004e8463c28d6ada669c475d19986a1a3689778c691717cff39a7531a"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/assets/charts/unified_task_model_radar.svg",
           "exists": true,
-          "bytes": 54717,
-          "sha256": "8fb6de7004e8463c28d6ada669c475d19986a1a3689778c691717cff39a7531a"
         },
         "hf_artifacts_card": {
           "path": "hf_artifacts:assets/charts/unified_task_model_radar.svg",
           "exists": true,
-          "bytes": 54717,
-          "sha256": "8fb6de7004e8463c28d6ada669c475d19986a1a3689778c691717cff39a7531a"
         },
         "hf_model": {
           "path": "hf_model:assets/charts/unified_task_model_radar.svg",
           "exists": true,
-          "bytes": 54717,
-          "sha256": "8fb6de7004e8463c28d6ada669c475d19986a1a3689778c691717cff39a7531a"
         }
       },
       "failures": []
@@ -4509,21 +4509,21 @@
       "local": {
         "path": "repo:scripts/omni/score_existing_model_output_task_probes.py",
         "exists": true,
-        "bytes": 57470,
-        "sha256": "9a0708885c1a22addc0bec199488021e277c7a7b7e23d6a37344fbbb3e5f1b96"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/omni/score_existing_model_output_task_probes.py",
           "exists": true,
-          "bytes": 57470,
-          "sha256": "9a0708885c1a22addc0bec199488021e277c7a7b7e23d6a37344fbbb3e5f1b96"
         },
         "hf_model": {
           "path": "hf_model:scripts/omni/score_existing_model_output_task_probes.py",
           "exists": true,
-          "bytes": 57470,
-          "sha256": "9a0708885c1a22addc0bec199488021e277c7a7b7e23d6a37344fbbb3e5f1b96"
         }
       },
       "failures": []
@@ -5034,21 +5034,21 @@
       "local": {
         "path": "repo:scripts/build_unified_task_model_radar.py",
         "exists": true,
-        "bytes": 56295,
-        "sha256": "237ef6a08dd81b9fcb0256f6f6b54394f74c14daea0cbba8f1726fdbbb7feff0"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_unified_task_model_radar.py",
           "exists": true,
-          "bytes": 56295,
-          "sha256": "237ef6a08dd81b9fcb0256f6f6b54394f74c14daea0cbba8f1726fdbbb7feff0"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_unified_task_model_radar.py",
           "exists": true,
-          "bytes": 56295,
-          "sha256": "237ef6a08dd81b9fcb0256f6f6b54394f74c14daea0cbba8f1726fdbbb7feff0"
         }
       },
       "failures": []
@@ -5464,38 +5464,38 @@
         "path": "repo:docs/index.html",
         "exists": true,
         "bytes": 254770,
-        "sha256": "9273d6b20b166a8d18bb30f13c9f15389a3808e8c71265d68dd9481911fc5d66"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:index.html",
           "exists": true,
           "bytes": 254770,
-          "sha256": "9273d6b20b166a8d18bb30f13c9f15389a3808e8c71265d68dd9481911fc5d66"
         },
         "hf_artifacts_root": {
           "path": "hf_artifacts:index.html",
           "exists": true,
           "bytes": 254770,
-          "sha256": "9273d6b20b166a8d18bb30f13c9f15389a3808e8c71265d68dd9481911fc5d66"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/index.html",
           "exists": true,
           "bytes": 254770,
-          "sha256": "9273d6b20b166a8d18bb30f13c9f15389a3808e8c71265d68dd9481911fc5d66"
         },
         "hf_model": {
           "path": "hf_model:index.html",
           "exists": true,
           "bytes": 254770,
-          "sha256": "9273d6b20b166a8d18bb30f13c9f15389a3808e8c71265d68dd9481911fc5d66"
         },
         "hf_model_docs": {
           "path": "hf_model:docs/index.html",
           "exists": true,
           "bytes": 254770,
-          "sha256": "9273d6b20b166a8d18bb30f13c9f15389a3808e8c71265d68dd9481911fc5d66"
         }
       },
       "failures": []
@@ -12510,21 +12510,21 @@
       "local": {
         "path": "repo:results/omni_finetune/model_output_probe_readiness/RUN_REPORT.md",
         "exists": true,
-        "bytes": 1047,
-        "sha256": "823b4dcc50bbaf8e5124abf3b9ec3c88dbfe7f5be1995769f7a537f31adfa31d"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_probe_readiness/RUN_REPORT.md",
           "exists": true,
-          "bytes": 1047,
-          "sha256": "823b4dcc50bbaf8e5124abf3b9ec3c88dbfe7f5be1995769f7a537f31adfa31d"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_probe_readiness/RUN_REPORT.md",
           "exists": true,
-          "bytes": 1047,
-          "sha256": "823b4dcc50bbaf8e5124abf3b9ec3c88dbfe7f5be1995769f7a537f31adfa31d"
         }
       },
       "failures": []
@@ -12535,21 +12535,21 @@
       "local": {
         "path": "repo:results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json",
         "exists": true,
-        "bytes": 5144,
-        "sha256": "86835a291c659d335ab5b51d09ea3f36710c14d23466a72d9c2a23137dc9ae8a"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json",
           "exists": true,
-          "bytes": 5144,
-          "sha256": "86835a291c659d335ab5b51d09ea3f36710c14d23466a72d9c2a23137dc9ae8a"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json",
           "exists": true,
-          "bytes": 5144,
-          "sha256": "86835a291c659d335ab5b51d09ea3f36710c14d23466a72d9c2a23137dc9ae8a"
         }
       },
       "failures": []
@@ -12560,21 +12560,96 @@
       "local": {
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
         "exists": true,
-        "bytes": 1421,
-        "sha256": "7f2156454573d3174ed5ddb029864fdf65214d9f8c09a151d0a3bd9a0ef86452"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
           "exists": true,
-          "bytes": 1421,
-          "sha256": "7f2156454573d3174ed5ddb029864fdf65214d9f8c09a151d0a3bd9a0ef86452"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
           "exists": true,
-          "bytes": 1421,
-          "sha256": "7f2156454573d3174ed5ddb029864fdf65214d9f8c09a151d0a3bd9a0ef86452"
         }
       },
       "failures": []
@@ -12586,20 +12661,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 21876,
-        "sha256": "a1d506a2c132d1d0e8bf31a9ba8618fec88bac75b056f99c50219514fdf3e22e"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 21876,
-          "sha256": "a1d506a2c132d1d0e8bf31a9ba8618fec88bac75b056f99c50219514fdf3e22e"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 21876,
-          "sha256": "a1d506a2c132d1d0e8bf31a9ba8618fec88bac75b056f99c50219514fdf3e22e"
         }
       },
       "failures": []
@@ -12661,20 +12736,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
         "exists": true,
         "bytes": 169938,
-        "sha256": "e5754d22070d2b25b60e03e1c9acde49daacce48b871c7f4520d130355c07eb1"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
           "exists": true,
           "bytes": 169938,
-          "sha256": "e5754d22070d2b25b60e03e1c9acde49daacce48b871c7f4520d130355c07eb1"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
           "exists": true,
           "bytes": 169938,
-          "sha256": "e5754d22070d2b25b60e03e1c9acde49daacce48b871c7f4520d130355c07eb1"
         }
       },
       "failures": []
@@ -12729,6 +12804,56 @@
       },
       "failures": []
     },
     {
       "name": "results/omni_finetune/model_output_task_probes_20260616/cosmos3_nano_future_window_source_window_map.jsonl",
       "status": "pass",
@@ -12786,20 +12911,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 7404,
-        "sha256": "fe39f6894d2b71411b4272e555d1801e55f3b4ed419e0d4cd19e0e4ed53f60f1"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 7404,
-          "sha256": "fe39f6894d2b71411b4272e555d1801e55f3b4ed419e0d4cd19e0e4ed53f60f1"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 7404,
-          "sha256": "fe39f6894d2b71411b4272e555d1801e55f3b4ed419e0d4cd19e0e4ed53f60f1"
         }
       },
       "failures": []
@@ -12861,20 +12986,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 10051,
-        "sha256": "b7773f4af1284dfb498c69ba174049056e038ce10465281b657223629d1dab5f"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 10051,
-          "sha256": "b7773f4af1284dfb498c69ba174049056e038ce10465281b657223629d1dab5f"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 10051,
-          "sha256": "b7773f4af1284dfb498c69ba174049056e038ce10465281b657223629d1dab5f"
         }
       },
       "failures": []
@@ -12936,20 +13061,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 1825,
-        "sha256": "4d04bb2e461a76d0a8f9ec37f6b73915c87544472406b3ca576cb791f100db92"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 1825,
-          "sha256": "4d04bb2e461a76d0a8f9ec37f6b73915c87544472406b3ca576cb791f100db92"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 1825,
-          "sha256": "4d04bb2e461a76d0a8f9ec37f6b73915c87544472406b3ca576cb791f100db92"
         }
       },
       "failures": []
@@ -12961,20 +13086,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 5594,
-        "sha256": "0f2b7b041ae7cd46b128427634d0328178303c1ede02643fb86a68933c149099"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 5594,
-          "sha256": "0f2b7b041ae7cd46b128427634d0328178303c1ede02643fb86a68933c149099"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 5594,
-          "sha256": "0f2b7b041ae7cd46b128427634d0328178303c1ede02643fb86a68933c149099"
         }
       },
       "failures": []
@@ -13036,20 +13161,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 2225,
-        "sha256": "f080f27c4457005ae1383d31cde22fc3c51b0c065eda3e0b5842135048254877"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2225,
-          "sha256": "f080f27c4457005ae1383d31cde22fc3c51b0c065eda3e0b5842135048254877"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2225,
-          "sha256": "f080f27c4457005ae1383d31cde22fc3c51b0c065eda3e0b5842135048254877"
         }
       },
       "failures": []
@@ -13085,21 +13210,21 @@
       "local": {
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/summary.json",
         "exists": true,
-        "bytes": 5278,
-        "sha256": "4a27f39807e8ac13771e2109584a8b0265961bcea88445058cbfe87265457d7f"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/summary.json",
           "exists": true,
-          "bytes": 5278,
-          "sha256": "4a27f39807e8ac13771e2109584a8b0265961bcea88445058cbfe87265457d7f"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/summary.json",
           "exists": true,
-          "bytes": 5278,
-          "sha256": "4a27f39807e8ac13771e2109584a8b0265961bcea88445058cbfe87265457d7f"
         }
       },
       "failures": []
@@ -13111,20 +13236,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 2398,
-        "sha256": "31b36b403d37ee21efed27e9c2c5c4b46806bb1caff9f9159c5914c6c104a61c"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2398,
-          "sha256": "31b36b403d37ee21efed27e9c2c5c4b46806bb1caff9f9159c5914c6c104a61c"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2398,
-          "sha256": "31b36b403d37ee21efed27e9c2c5c4b46806bb1caff9f9159c5914c6c104a61c"
         }
       },
       "failures": []
@@ -13161,20 +13286,20 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 2049,
-        "sha256": "6eae6e5b12a4073e98531046451c8ef332f0ca469a741aab09e94c952dc754cb"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 2049,
-          "sha256": "6eae6e5b12a4073e98531046451c8ef332f0ca469a741aab09e94c952dc754cb"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 2049,
-          "sha256": "6eae6e5b12a4073e98531046451c8ef332f0ca469a741aab09e94c952dc754cb"
         }
       },
       "failures": []
@@ -20667,15 +20792,72 @@
       "local": {
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
         "exists": true,
-        "bytes": 1421,
-        "sha256": "7f2156454573d3174ed5ddb029864fdf65214d9f8c09a151d0a3bd9a0ef86452"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
           "exists": true,
-          "bytes": 1421,
-          "sha256": "7f2156454573d3174ed5ddb029864fdf65214d9f8c09a151d0a3bd9a0ef86452"
         }
       },
       "failures": []
@@ -20687,14 +20869,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 21876,
-        "sha256": "a1d506a2c132d1d0e8bf31a9ba8618fec88bac75b056f99c50219514fdf3e22e"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 21876,
-          "sha256": "a1d506a2c132d1d0e8bf31a9ba8618fec88bac75b056f99c50219514fdf3e22e"
         }
       },
       "failures": []
@@ -20744,14 +20926,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
         "exists": true,
         "bytes": 169938,
-        "sha256": "e5754d22070d2b25b60e03e1c9acde49daacce48b871c7f4520d130355c07eb1"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
           "exists": true,
           "bytes": 169938,
-          "sha256": "e5754d22070d2b25b60e03e1c9acde49daacce48b871c7f4520d130355c07eb1"
         }
       },
       "failures": []
@@ -20794,6 +20976,44 @@
       },
       "failures": []
     },
     {
       "name": "space_results/omni_finetune/model_output_task_probes_20260616/cosmos3_nano_future_window_source_window_map.jsonl",
       "status": "pass",
@@ -20839,14 +21059,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 7404,
-        "sha256": "fe39f6894d2b71411b4272e555d1801e55f3b4ed419e0d4cd19e0e4ed53f60f1"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 7404,
-          "sha256": "fe39f6894d2b71411b4272e555d1801e55f3b4ed419e0d4cd19e0e4ed53f60f1"
         }
       },
       "failures": []
@@ -20896,14 +21116,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 10051,
-        "sha256": "b7773f4af1284dfb498c69ba174049056e038ce10465281b657223629d1dab5f"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 10051,
-          "sha256": "b7773f4af1284dfb498c69ba174049056e038ce10465281b657223629d1dab5f"
         }
       },
       "failures": []
@@ -20953,14 +21173,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 1825,
-        "sha256": "4d04bb2e461a76d0a8f9ec37f6b73915c87544472406b3ca576cb791f100db92"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 1825,
-          "sha256": "4d04bb2e461a76d0a8f9ec37f6b73915c87544472406b3ca576cb791f100db92"
         }
       },
       "failures": []
@@ -20972,14 +21192,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 5594,
-        "sha256": "0f2b7b041ae7cd46b128427634d0328178303c1ede02643fb86a68933c149099"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 5594,
-          "sha256": "0f2b7b041ae7cd46b128427634d0328178303c1ede02643fb86a68933c149099"
         }
       },
       "failures": []
@@ -21029,14 +21249,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 2225,
-        "sha256": "f080f27c4457005ae1383d31cde22fc3c51b0c065eda3e0b5842135048254877"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2225,
-          "sha256": "f080f27c4457005ae1383d31cde22fc3c51b0c065eda3e0b5842135048254877"
         }
       },
       "failures": []
@@ -21066,15 +21286,15 @@
       "local": {
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/summary.json",
         "exists": true,
-        "bytes": 5278,
-        "sha256": "4a27f39807e8ac13771e2109584a8b0265961bcea88445058cbfe87265457d7f"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/summary.json",
           "exists": true,
-          "bytes": 5278,
-          "sha256": "4a27f39807e8ac13771e2109584a8b0265961bcea88445058cbfe87265457d7f"
         }
       },
       "failures": []
@@ -21086,14 +21306,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 2398,
-        "sha256": "31b36b403d37ee21efed27e9c2c5c4b46806bb1caff9f9159c5914c6c104a61c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2398,
-          "sha256": "31b36b403d37ee21efed27e9c2c5c4b46806bb1caff9f9159c5914c6c104a61c"
         }
       },
       "failures": []
@@ -21124,14 +21344,14 @@
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 2049,
-        "sha256": "6eae6e5b12a4073e98531046451c8ef332f0ca469a741aab09e94c952dc754cb"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 2049,
-          "sha256": "6eae6e5b12a4073e98531046451c8ef332f0ca469a741aab09e94c952dc754cb"
         }
       },
       "failures": []
@@ -22710,27 +22930,27 @@
       "local": {
         "path": "repo:TASK_METHOD_20_GAP_AUDIT.md",
         "exists": true,
-        "bytes": 8787,
-        "sha256": "8fd8555f75cd96c31346618ab622e24655027cea5c52e8a09a7a53793ef7b72b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:TASK_METHOD_20_GAP_AUDIT.md",
           "exists": true,
-          "bytes": 8787,
-          "sha256": "8fd8555f75cd96c31346618ab622e24655027cea5c52e8a09a7a53793ef7b72b"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:TASK_METHOD_20_GAP_AUDIT.md",
           "exists": true,
-          "bytes": 8787,
-          "sha256": "8fd8555f75cd96c31346618ab622e24655027cea5c52e8a09a7a53793ef7b72b"
         },
         "hf_model": {
           "path": "hf_model:TASK_METHOD_20_GAP_AUDIT.md",
           "exists": true,
-          "bytes": 8787,
-          "sha256": "8fd8555f75cd96c31346618ab622e24655027cea5c52e8a09a7a53793ef7b72b"
         }
       },
       "failures": []
@@ -22741,27 +22961,27 @@
       "local": {
         "path": "repo:TASK_METHOD_20_RESULT_MATRIX.md",
         "exists": true,
-        "bytes": 3892,
-        "sha256": "590d9362624d64b81b1b19638c46faa7969730d40415e4825abd2ff09d76340a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:TASK_METHOD_20_RESULT_MATRIX.md",
           "exists": true,
-          "bytes": 3892,
-          "sha256": "590d9362624d64b81b1b19638c46faa7969730d40415e4825abd2ff09d76340a"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:TASK_METHOD_20_RESULT_MATRIX.md",
           "exists": true,
-          "bytes": 3892,
-          "sha256": "590d9362624d64b81b1b19638c46faa7969730d40415e4825abd2ff09d76340a"
         },
         "hf_model": {
           "path": "hf_model:TASK_METHOD_20_RESULT_MATRIX.md",
           "exists": true,
-          "bytes": 3892,
-          "sha256": "590d9362624d64b81b1b19638c46faa7969730d40415e4825abd2ff09d76340a"
         }
       },
       "failures": []

 {
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:59:08+00:00",
   "hf_root": "hf_publish",
   "summary": {
+    "group_count": 865,
     "failure_count": 0,
     "failures_by_surface": {}
   },
       "local": {
         "path": "repo:docs/data/artifact_index.json",
         "exists": true,
+        "bytes": 116644,
+        "sha256": "af87562c9963c897035e5d2a2f562423989a750c24809438f41b982986b7c658"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/artifact_index.json",
           "exists": true,
+          "bytes": 116644,
+          "sha256": "af87562c9963c897035e5d2a2f562423989a750c24809438f41b982986b7c658"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/artifact_index.json",
           "exists": true,
+          "bytes": 116644,
+          "sha256": "af87562c9963c897035e5d2a2f562423989a750c24809438f41b982986b7c658"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/artifact_index.json",
           "exists": true,
+          "bytes": 116644,
+          "sha256": "af87562c9963c897035e5d2a2f562423989a750c24809438f41b982986b7c658"
         },
         "hf_model_data": {
           "path": "hf_model:data/artifact_index.json",
           "exists": true,
+          "bytes": 116644,
+          "sha256": "af87562c9963c897035e5d2a2f562423989a750c24809438f41b982986b7c658"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/artifact_index.json",
           "exists": true,
+          "bytes": 116644,
+          "sha256": "af87562c9963c897035e5d2a2f562423989a750c24809438f41b982986b7c658"
         },
         "hf_model": {
           "path": "hf_model:metrics/artifact_index.json",
           "exists": true,
+          "bytes": 116644,
+          "sha256": "af87562c9963c897035e5d2a2f562423989a750c24809438f41b982986b7c658"
         }
       },
       "failures": []
       "local": {
         "path": "repo:docs/data/publication_audit.json",
         "exists": true,
+        "bytes": 9713,
+        "sha256": "33c3601546d5ff2e7c093071d8db2331d73097ac5360b60bbebb8902c2a6dda5"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/publication_audit.json",
           "exists": true,
+          "bytes": 9713,
+          "sha256": "33c3601546d5ff2e7c093071d8db2331d73097ac5360b60bbebb8902c2a6dda5"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/publication_audit.json",
           "exists": true,
+          "bytes": 9713,
+          "sha256": "33c3601546d5ff2e7c093071d8db2331d73097ac5360b60bbebb8902c2a6dda5"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/publication_audit.json",
           "exists": true,
+          "bytes": 9713,
+          "sha256": "33c3601546d5ff2e7c093071d8db2331d73097ac5360b60bbebb8902c2a6dda5"
         },
         "hf_model_data": {
           "path": "hf_model:data/publication_audit.json",
           "exists": true,
+          "bytes": 9713,
+          "sha256": "33c3601546d5ff2e7c093071d8db2331d73097ac5360b60bbebb8902c2a6dda5"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/publication_audit.json",
           "exists": true,
+          "bytes": 9713,
+          "sha256": "33c3601546d5ff2e7c093071d8db2331d73097ac5360b60bbebb8902c2a6dda5"
         },
         "hf_model": {
           "path": "hf_model:metrics/publication_audit.json",
           "exists": true,
+          "bytes": 9713,
+          "sha256": "33c3601546d5ff2e7c093071d8db2331d73097ac5360b60bbebb8902c2a6dda5"
         }
       },
       "failures": []
         "path": "repo:docs/data/public_surface_qa.json",
         "exists": true,
         "bytes": 7126,
+        "sha256": "0fd0ee7e07b83212d50cf95fa1e97c78baa0020ad26148122f28649b6829f082"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
+          "sha256": "0fd0ee7e07b83212d50cf95fa1e97c78baa0020ad26148122f28649b6829f082"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
+          "sha256": "0fd0ee7e07b83212d50cf95fa1e97c78baa0020ad26148122f28649b6829f082"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
+          "sha256": "0fd0ee7e07b83212d50cf95fa1e97c78baa0020ad26148122f28649b6829f082"
         },
         "hf_model_data": {
           "path": "hf_model:data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
+          "sha256": "0fd0ee7e07b83212d50cf95fa1e97c78baa0020ad26148122f28649b6829f082"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
+          "sha256": "0fd0ee7e07b83212d50cf95fa1e97c78baa0020ad26148122f28649b6829f082"
         },
         "hf_model": {
           "path": "hf_model:metrics/public_surface_qa.json",
           "exists": true,
           "bytes": 7126,
+          "sha256": "0fd0ee7e07b83212d50cf95fa1e97c78baa0020ad26148122f28649b6829f082"
         }
       },
       "failures": []
         "path": "repo:docs/data/quality_gates.json",
         "exists": true,
         "bytes": 8100,
+        "sha256": "b88b46bd9b16d4afaf4836d0acff208b46559a493cfcffa881f1528fa282292c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
+          "sha256": "b88b46bd9b16d4afaf4836d0acff208b46559a493cfcffa881f1528fa282292c"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
+          "sha256": "b88b46bd9b16d4afaf4836d0acff208b46559a493cfcffa881f1528fa282292c"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
+          "sha256": "b88b46bd9b16d4afaf4836d0acff208b46559a493cfcffa881f1528fa282292c"
         },
         "hf_model_data": {
           "path": "hf_model:data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
+          "sha256": "b88b46bd9b16d4afaf4836d0acff208b46559a493cfcffa881f1528fa282292c"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/quality_gates.json",
           "exists": true,
           "bytes": 8100,
+          "sha256": "b88b46bd9b16d4afaf4836d0acff208b46559a493cfcffa881f1528fa282292c"
         },
         "hf_model": {
           "path": "hf_model:metrics/quality_gates.json",
           "exists": true,
           "bytes": 8100,
+          "sha256": "b88b46bd9b16d4afaf4836d0acff208b46559a493cfcffa881f1528fa282292c"
         }
       },
       "failures": []
         "path": "repo:docs/data/scope_claims_audit.json",
         "exists": true,
         "bytes": 21630,
+        "sha256": "a4d8a83955779c23e3cf985a2cbbc49fd544c85bf8495c1ff7d37d9630fcc479"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
+          "sha256": "a4d8a83955779c23e3cf985a2cbbc49fd544c85bf8495c1ff7d37d9630fcc479"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
+          "sha256": "a4d8a83955779c23e3cf985a2cbbc49fd544c85bf8495c1ff7d37d9630fcc479"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
+          "sha256": "a4d8a83955779c23e3cf985a2cbbc49fd544c85bf8495c1ff7d37d9630fcc479"
         },
         "hf_model_data": {
           "path": "hf_model:data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
+          "sha256": "a4d8a83955779c23e3cf985a2cbbc49fd544c85bf8495c1ff7d37d9630fcc479"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
+          "sha256": "a4d8a83955779c23e3cf985a2cbbc49fd544c85bf8495c1ff7d37d9630fcc479"
         },
         "hf_model": {
           "path": "hf_model:metrics/scope_claims_audit.json",
           "exists": true,
           "bytes": 21630,
+          "sha256": "a4d8a83955779c23e3cf985a2cbbc49fd544c85bf8495c1ff7d37d9630fcc479"
         }
       },
       "failures": []
         "path": "repo:docs/data/source_alignment_audit.json",
         "exists": true,
         "bytes": 4432,
+        "sha256": "d51f321ea35f0e57aeaaa41dad9af0cccdc1fde2c61e9559cc374a1e5fd85f4a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
+          "sha256": "d51f321ea35f0e57aeaaa41dad9af0cccdc1fde2c61e9559cc374a1e5fd85f4a"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
+          "sha256": "d51f321ea35f0e57aeaaa41dad9af0cccdc1fde2c61e9559cc374a1e5fd85f4a"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
+          "sha256": "d51f321ea35f0e57aeaaa41dad9af0cccdc1fde2c61e9559cc374a1e5fd85f4a"
         },
         "hf_model_data": {
           "path": "hf_model:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
+          "sha256": "d51f321ea35f0e57aeaaa41dad9af0cccdc1fde2c61e9559cc374a1e5fd85f4a"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
+          "sha256": "d51f321ea35f0e57aeaaa41dad9af0cccdc1fde2c61e9559cc374a1e5fd85f4a"
         },
         "hf_model": {
           "path": "hf_model:metrics/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
+          "sha256": "d51f321ea35f0e57aeaaa41dad9af0cccdc1fde2c61e9559cc374a1e5fd85f4a"
         }
       },
       "failures": []
         "path": "repo:docs/data/single_episode_task_model_radar.json",
         "exists": true,
         "bytes": 51064,
+        "sha256": "9d42b045d833e50a2bfa8905203150af1f28ea4736d22edb0fe27e39eedf4574"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
+          "sha256": "9d42b045d833e50a2bfa8905203150af1f28ea4736d22edb0fe27e39eedf4574"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
+          "sha256": "9d42b045d833e50a2bfa8905203150af1f28ea4736d22edb0fe27e39eedf4574"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
+          "sha256": "9d42b045d833e50a2bfa8905203150af1f28ea4736d22edb0fe27e39eedf4574"
         },
         "hf_model_data": {
           "path": "hf_model:data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
+          "sha256": "9d42b045d833e50a2bfa8905203150af1f28ea4736d22edb0fe27e39eedf4574"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
+          "sha256": "9d42b045d833e50a2bfa8905203150af1f28ea4736d22edb0fe27e39eedf4574"
         },
         "hf_model": {
           "path": "hf_model:metrics/single_episode_task_model_radar.json",
           "exists": true,
           "bytes": 51064,
+          "sha256": "9d42b045d833e50a2bfa8905203150af1f28ea4736d22edb0fe27e39eedf4574"
         }
       },
       "failures": []
       "local": {
         "path": "repo:docs/data/episode128_task_model_radar.json",
         "exists": true,
+        "bytes": 185314,
+        "sha256": "5c61c8f9a4a9a7b6bdc65e39ef5fda6117df48412afe260c0be4da501bd7e8db"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/episode128_task_model_radar.json",
           "exists": true,
+          "bytes": 185314,
+          "sha256": "5c61c8f9a4a9a7b6bdc65e39ef5fda6117df48412afe260c0be4da501bd7e8db"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/episode128_task_model_radar.json",
           "exists": true,
+          "bytes": 185314,
+          "sha256": "5c61c8f9a4a9a7b6bdc65e39ef5fda6117df48412afe260c0be4da501bd7e8db"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/episode128_task_model_radar.json",
           "exists": true,
+          "bytes": 185314,
+          "sha256": "5c61c8f9a4a9a7b6bdc65e39ef5fda6117df48412afe260c0be4da501bd7e8db"
         },
         "hf_model_data": {
           "path": "hf_model:data/episode128_task_model_radar.json",
           "exists": true,
+          "bytes": 185314,
+          "sha256": "5c61c8f9a4a9a7b6bdc65e39ef5fda6117df48412afe260c0be4da501bd7e8db"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/episode128_task_model_radar.json",
           "exists": true,
+          "bytes": 185314,
+          "sha256": "5c61c8f9a4a9a7b6bdc65e39ef5fda6117df48412afe260c0be4da501bd7e8db"
         },
         "hf_model": {
           "path": "hf_model:metrics/episode128_task_model_radar.json",
           "exists": true,
+          "bytes": 185314,
+          "sha256": "5c61c8f9a4a9a7b6bdc65e39ef5fda6117df48412afe260c0be4da501bd7e8db"
         }
       },
       "failures": []
         "path": "repo:docs/data/task_surface_integrity.json",
         "exists": true,
         "bytes": 45779,
+        "sha256": "53473545e9177adee1eb5c4245008c7b101a7b16232dce9dd38e509ae8c867dd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
+          "sha256": "53473545e9177adee1eb5c4245008c7b101a7b16232dce9dd38e509ae8c867dd"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
+          "sha256": "53473545e9177adee1eb5c4245008c7b101a7b16232dce9dd38e509ae8c867dd"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
+          "sha256": "53473545e9177adee1eb5c4245008c7b101a7b16232dce9dd38e509ae8c867dd"
         },
         "hf_model_data": {
           "path": "hf_model:data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
+          "sha256": "53473545e9177adee1eb5c4245008c7b101a7b16232dce9dd38e509ae8c867dd"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
+          "sha256": "53473545e9177adee1eb5c4245008c7b101a7b16232dce9dd38e509ae8c867dd"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
+          "sha256": "53473545e9177adee1eb5c4245008c7b101a7b16232dce9dd38e509ae8c867dd"
         }
       },
       "failures": []
       "local": {
         "path": "repo:docs/data/task_method_20_result_matrix.json",
         "exists": true,
+        "bytes": 128805,
+        "sha256": "d5572768c6fb671eadf87870d20c8645dfae384f862aa63e929b2db50a5cc51a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_method_20_result_matrix.json",
           "exists": true,
+          "bytes": 128805,
+          "sha256": "d5572768c6fb671eadf87870d20c8645dfae384f862aa63e929b2db50a5cc51a"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/task_method_20_result_matrix.json",
           "exists": true,
+          "bytes": 128805,
+          "sha256": "d5572768c6fb671eadf87870d20c8645dfae384f862aa63e929b2db50a5cc51a"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_method_20_result_matrix.json",
           "exists": true,
+          "bytes": 128805,
+          "sha256": "d5572768c6fb671eadf87870d20c8645dfae384f862aa63e929b2db50a5cc51a"
         },
         "hf_model_data": {
           "path": "hf_model:data/task_method_20_result_matrix.json",
           "exists": true,
+          "bytes": 128805,
+          "sha256": "d5572768c6fb671eadf87870d20c8645dfae384f862aa63e929b2db50a5cc51a"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/task_method_20_result_matrix.json",
           "exists": true,
+          "bytes": 128805,
+          "sha256": "d5572768c6fb671eadf87870d20c8645dfae384f862aa63e929b2db50a5cc51a"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_method_20_result_matrix.json",
           "exists": true,
+          "bytes": 128805,
+          "sha256": "d5572768c6fb671eadf87870d20c8645dfae384f862aa63e929b2db50a5cc51a"
         }
       },
       "failures": []
       "local": {
         "path": "repo:docs/data/task_method_20_gap_audit.json",
         "exists": true,
+        "bytes": 28259,
+        "sha256": "860dea3d3ce2f2c2146c97918a5deca05ae315238b584c10e53f74a9d2c26dcd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_method_20_gap_audit.json",
           "exists": true,
+          "bytes": 28259,
+          "sha256": "860dea3d3ce2f2c2146c97918a5deca05ae315238b584c10e53f74a9d2c26dcd"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/task_method_20_gap_audit.json",
           "exists": true,
+          "bytes": 28259,
+          "sha256": "860dea3d3ce2f2c2146c97918a5deca05ae315238b584c10e53f74a9d2c26dcd"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_method_20_gap_audit.json",
           "exists": true,
+          "bytes": 28259,
+          "sha256": "860dea3d3ce2f2c2146c97918a5deca05ae315238b584c10e53f74a9d2c26dcd"
         },
         "hf_model_data": {
           "path": "hf_model:data/task_method_20_gap_audit.json",
           "exists": true,
+          "bytes": 28259,
+          "sha256": "860dea3d3ce2f2c2146c97918a5deca05ae315238b584c10e53f74a9d2c26dcd"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/task_method_20_gap_audit.json",
           "exists": true,
+          "bytes": 28259,
+          "sha256": "860dea3d3ce2f2c2146c97918a5deca05ae315238b584c10e53f74a9d2c26dcd"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_method_20_gap_audit.json",
           "exists": true,
+          "bytes": 28259,
+          "sha256": "860dea3d3ce2f2c2146c97918a5deca05ae315238b584c10e53f74a9d2c26dcd"
         }
       },
       "failures": []
       "local": {
         "path": "repo:docs/data/unified_task_model_radar.json",
         "exists": true,
+        "bytes": 229168,
+        "sha256": "64c7c10f3ce663178f2eea4740bb45de0f94b1246448f14f5521d293be916e77"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/unified_task_model_radar.json",
           "exists": true,
+          "bytes": 229168,
+          "sha256": "64c7c10f3ce663178f2eea4740bb45de0f94b1246448f14f5521d293be916e77"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/unified_task_model_radar.json",
           "exists": true,
+          "bytes": 229168,
+          "sha256": "64c7c10f3ce663178f2eea4740bb45de0f94b1246448f14f5521d293be916e77"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/unified_task_model_radar.json",
           "exists": true,
+          "bytes": 229168,
+          "sha256": "64c7c10f3ce663178f2eea4740bb45de0f94b1246448f14f5521d293be916e77"
         },
         "hf_model_data": {
           "path": "hf_model:data/unified_task_model_radar.json",
           "exists": true,
+          "bytes": 229168,
+          "sha256": "64c7c10f3ce663178f2eea4740bb45de0f94b1246448f14f5521d293be916e77"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/unified_task_model_radar.json",
           "exists": true,
+          "bytes": 229168,
+          "sha256": "64c7c10f3ce663178f2eea4740bb45de0f94b1246448f14f5521d293be916e77"
         },
         "hf_model": {
           "path": "hf_model:metrics/unified_task_model_radar.json",
           "exists": true,
+          "bytes": 229168,
+          "sha256": "64c7c10f3ce663178f2eea4740bb45de0f94b1246448f14f5521d293be916e77"
         }
       },
       "failures": []
       "local": {
         "path": "repo:docs/data/website_integrity.json",
         "exists": true,
+        "bytes": 20022,
+        "sha256": "4b46dacca993c2a0645cc79518d40b4e57198fe304daa322d0daa5a268043538"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/website_integrity.json",
           "exists": true,
+          "bytes": 20022,
+          "sha256": "4b46dacca993c2a0645cc79518d40b4e57198fe304daa322d0daa5a268043538"
         },
         "hf_artifacts_data": {
           "path": "hf_artifacts:data/website_integrity.json",
           "exists": true,
+          "bytes": 20022,
+          "sha256": "4b46dacca993c2a0645cc79518d40b4e57198fe304daa322d0daa5a268043538"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/website_integrity.json",
           "exists": true,
+          "bytes": 20022,
+          "sha256": "4b46dacca993c2a0645cc79518d40b4e57198fe304daa322d0daa5a268043538"
         },
         "hf_model_data": {
           "path": "hf_model:data/website_integrity.json",
           "exists": true,
+          "bytes": 20022,
+          "sha256": "4b46dacca993c2a0645cc79518d40b4e57198fe304daa322d0daa5a268043538"
         },
         "hf_model_docs_data": {
           "path": "hf_model:docs/data/website_integrity.json",
           "exists": true,
+          "bytes": 20022,
+          "sha256": "4b46dacca993c2a0645cc79518d40b4e57198fe304daa322d0daa5a268043538"
         },
         "hf_model": {
           "path": "hf_model:metrics/website_integrity.json",
           "exists": true,
+          "bytes": 20022,
+          "sha256": "4b46dacca993c2a0645cc79518d40b4e57198fe304daa322d0daa5a268043538"
         }
       },
       "failures": []
       "local": {
         "path": "repo:docs/assets/charts/episode128_task_model_radar.svg",
         "exists": true,
+        "bytes": 48923,
+        "sha256": "75c389122a413548bdc26fd05cbbad2a85015ebd222ba1a23571f3ff4f0eddcf"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:assets/charts/episode128_task_model_radar.svg",
           "exists": true,
+          "bytes": 48923,
+          "sha256": "75c389122a413548bdc26fd05cbbad2a85015ebd222ba1a23571f3ff4f0eddcf"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/assets/charts/episode128_task_model_radar.svg",
           "exists": true,
+          "bytes": 48923,
+          "sha256": "75c389122a413548bdc26fd05cbbad2a85015ebd222ba1a23571f3ff4f0eddcf"
         },
         "hf_artifacts_card": {
           "path": "hf_artifacts:assets/charts/episode128_task_model_radar.svg",
           "exists": true,
+          "bytes": 48923,
+          "sha256": "75c389122a413548bdc26fd05cbbad2a85015ebd222ba1a23571f3ff4f0eddcf"
         },
         "hf_model": {
           "path": "hf_model:assets/charts/episode128_task_model_radar.svg",
           "exists": true,
+          "bytes": 48923,
+          "sha256": "75c389122a413548bdc26fd05cbbad2a85015ebd222ba1a23571f3ff4f0eddcf"
         }
       },
       "failures": []
       "local": {
         "path": "repo:docs/assets/charts/unified_task_model_radar.svg",
         "exists": true,
+        "bytes": 54936,
+        "sha256": "f29db4099d60c093ebd092bee37a2fc1338bbff17d7e52b8a1b27ae6fc06727a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:assets/charts/unified_task_model_radar.svg",
           "exists": true,
+          "bytes": 54936,
+          "sha256": "f29db4099d60c093ebd092bee37a2fc1338bbff17d7e52b8a1b27ae6fc06727a"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/assets/charts/unified_task_model_radar.svg",
           "exists": true,
+          "bytes": 54936,
+          "sha256": "f29db4099d60c093ebd092bee37a2fc1338bbff17d7e52b8a1b27ae6fc06727a"
         },
         "hf_artifacts_card": {
           "path": "hf_artifacts:assets/charts/unified_task_model_radar.svg",
           "exists": true,
+          "bytes": 54936,
+          "sha256": "f29db4099d60c093ebd092bee37a2fc1338bbff17d7e52b8a1b27ae6fc06727a"
         },
         "hf_model": {
           "path": "hf_model:assets/charts/unified_task_model_radar.svg",
           "exists": true,
+          "bytes": 54936,
+          "sha256": "f29db4099d60c093ebd092bee37a2fc1338bbff17d7e52b8a1b27ae6fc06727a"
         }
       },
       "failures": []
       "local": {
         "path": "repo:scripts/omni/score_existing_model_output_task_probes.py",
         "exists": true,
+        "bytes": 69411,
+        "sha256": "a78889e5225405fcbb0156aa8c63cbbe79022a12283d0b398c5f2bb015175fe7"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/omni/score_existing_model_output_task_probes.py",
           "exists": true,
+          "bytes": 69411,
+          "sha256": "a78889e5225405fcbb0156aa8c63cbbe79022a12283d0b398c5f2bb015175fe7"
         },
         "hf_model": {
           "path": "hf_model:scripts/omni/score_existing_model_output_task_probes.py",
           "exists": true,
+          "bytes": 69411,
+          "sha256": "a78889e5225405fcbb0156aa8c63cbbe79022a12283d0b398c5f2bb015175fe7"
         }
       },
       "failures": []
       "local": {
         "path": "repo:scripts/build_unified_task_model_radar.py",
         "exists": true,
+        "bytes": 57104,
+        "sha256": "805cfce89529fec73edd3e436c1a76afae46cebe31ba5e441b6dfb5cc142c935"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_unified_task_model_radar.py",
           "exists": true,
+          "bytes": 57104,
+          "sha256": "805cfce89529fec73edd3e436c1a76afae46cebe31ba5e441b6dfb5cc142c935"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_unified_task_model_radar.py",
           "exists": true,
+          "bytes": 57104,
+          "sha256": "805cfce89529fec73edd3e436c1a76afae46cebe31ba5e441b6dfb5cc142c935"
         }
       },
       "failures": []
         "path": "repo:docs/index.html",
         "exists": true,
         "bytes": 254770,
+        "sha256": "55fbaf659f3ffe067603938476f895c52782ab8978c701545b6ed53dc23ec8bd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:index.html",
           "exists": true,
           "bytes": 254770,
+          "sha256": "55fbaf659f3ffe067603938476f895c52782ab8978c701545b6ed53dc23ec8bd"
         },
         "hf_artifacts_root": {
           "path": "hf_artifacts:index.html",
           "exists": true,
           "bytes": 254770,
+          "sha256": "55fbaf659f3ffe067603938476f895c52782ab8978c701545b6ed53dc23ec8bd"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/index.html",
           "exists": true,
           "bytes": 254770,
+          "sha256": "55fbaf659f3ffe067603938476f895c52782ab8978c701545b6ed53dc23ec8bd"
         },
         "hf_model": {
           "path": "hf_model:index.html",
           "exists": true,
           "bytes": 254770,
+          "sha256": "55fbaf659f3ffe067603938476f895c52782ab8978c701545b6ed53dc23ec8bd"
         },
         "hf_model_docs": {
           "path": "hf_model:docs/index.html",
           "exists": true,
           "bytes": 254770,
+          "sha256": "55fbaf659f3ffe067603938476f895c52782ab8978c701545b6ed53dc23ec8bd"
         }
       },
       "failures": []
       "local": {
         "path": "repo:results/omni_finetune/model_output_probe_readiness/RUN_REPORT.md",
         "exists": true,
+        "bytes": 1048,
+        "sha256": "5744154e5c35e1c842826ea9bf33a800517e90572a562449e56d638c3211f064"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_probe_readiness/RUN_REPORT.md",
           "exists": true,
+          "bytes": 1048,
+          "sha256": "5744154e5c35e1c842826ea9bf33a800517e90572a562449e56d638c3211f064"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_probe_readiness/RUN_REPORT.md",
           "exists": true,
+          "bytes": 1048,
+          "sha256": "5744154e5c35e1c842826ea9bf33a800517e90572a562449e56d638c3211f064"
         }
       },
       "failures": []
       "local": {
         "path": "repo:results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json",
         "exists": true,
+        "bytes": 4977,
+        "sha256": "b9caf6d32bb801d845b57b4450b92c10a4c3f9d14890f6a209e5a52aede92568"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json",
           "exists": true,
+          "bytes": 4977,
+          "sha256": "b9caf6d32bb801d845b57b4450b92c10a4c3f9d14890f6a209e5a52aede92568"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json",
           "exists": true,
+          "bytes": 4977,
+          "sha256": "b9caf6d32bb801d845b57b4450b92c10a4c3f9d14890f6a209e5a52aede92568"
         }
       },
       "failures": []
       "local": {
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
         "exists": true,
+        "bytes": 1512,
+        "sha256": "12ea92277ddc22fdd29adc3c7288ca0e0ebec3be3ffdc47f1ff7c4ec075dd015"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
           "exists": true,
+          "bytes": 1512,
+          "sha256": "12ea92277ddc22fdd29adc3c7288ca0e0ebec3be3ffdc47f1ff7c4ec075dd015"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
           "exists": true,
+          "bytes": 1512,
+          "sha256": "12ea92277ddc22fdd29adc3c7288ca0e0ebec3be3ffdc47f1ff7c4ec075dd015"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
+        "exists": true,
+        "bytes": 12958,
+        "sha256": "0ea0cacdea44ed714fd47166a999606d76200a30453a39142554cc6bf1b12269"
+      },
+      "mirrors": {
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
+          "exists": true,
+          "bytes": 12958,
+          "sha256": "0ea0cacdea44ed714fd47166a999606d76200a30453a39142554cc6bf1b12269"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
+          "exists": true,
+          "bytes": 12958,
+          "sha256": "0ea0cacdea44ed714fd47166a999606d76200a30453a39142554cc6bf1b12269"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv",
+        "exists": true,
+        "bytes": 12133,
+        "sha256": "579c606563073db2ff09a636b294b41aad1ec8d0fa8ae19b8649d3273cf88a51"
+      },
+      "mirrors": {
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 12133,
+          "sha256": "579c606563073db2ff09a636b294b41aad1ec8d0fa8ae19b8649d3273cf88a51"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 12133,
+          "sha256": "579c606563073db2ff09a636b294b41aad1ec8d0fa8ae19b8649d3273cf88a51"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv",
+        "exists": true,
+        "bytes": 125863,
+        "sha256": "e8a55a2b42315e2febea992138f9ae31a5543c81219127a25f00f90cdc2440ba"
+      },
+      "mirrors": {
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv",
+          "exists": true,
+          "bytes": 125863,
+          "sha256": "e8a55a2b42315e2febea992138f9ae31a5543c81219127a25f00f90cdc2440ba"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv",
+          "exists": true,
+          "bytes": 125863,
+          "sha256": "e8a55a2b42315e2febea992138f9ae31a5543c81219127a25f00f90cdc2440ba"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 21876,
+        "sha256": "af47c1007e54e06553578a2fee4c9c6a9cbd730bbef0330a32ffd846bb364f14"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 21876,
+          "sha256": "af47c1007e54e06553578a2fee4c9c6a9cbd730bbef0330a32ffd846bb364f14"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 21876,
+          "sha256": "af47c1007e54e06553578a2fee4c9c6a9cbd730bbef0330a32ffd846bb364f14"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
         "exists": true,
         "bytes": 169938,
+        "sha256": "d1f1c299960ad62c684fe8b798467b49157b75f6bc89e8a10066cda71b66e305"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
           "exists": true,
           "bytes": 169938,
+          "sha256": "d1f1c299960ad62c684fe8b798467b49157b75f6bc89e8a10066cda71b66e305"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
           "exists": true,
           "bytes": 169938,
+          "sha256": "d1f1c299960ad62c684fe8b798467b49157b75f6bc89e8a10066cda71b66e305"
         }
       },
       "failures": []
       },
       "failures": []
     },
+    {
+      "name": "results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
+        "exists": true,
+        "bytes": 1738,
+        "sha256": "990efd574e384b785187d442858f17d6da1342ff4ab435ee569e5aa1f6faebc3"
+      },
+      "mirrors": {
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
+          "exists": true,
+          "bytes": 1738,
+          "sha256": "990efd574e384b785187d442858f17d6da1342ff4ab435ee569e5aa1f6faebc3"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
+          "exists": true,
+          "bytes": 1738,
+          "sha256": "990efd574e384b785187d442858f17d6da1342ff4ab435ee569e5aa1f6faebc3"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv",
+        "exists": true,
+        "bytes": 57340,
+        "sha256": "4216a73df04338edf271aaeafe66b041675726d21909e1eb25e7face575c567f"
+      },
+      "mirrors": {
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv",
+          "exists": true,
+          "bytes": 57340,
+          "sha256": "4216a73df04338edf271aaeafe66b041675726d21909e1eb25e7face575c567f"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv",
+          "exists": true,
+          "bytes": 57340,
+          "sha256": "4216a73df04338edf271aaeafe66b041675726d21909e1eb25e7face575c567f"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "results/omni_finetune/model_output_task_probes_20260616/cosmos3_nano_future_window_source_window_map.jsonl",
       "status": "pass",
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 7404,
+        "sha256": "4d6781bdd3baffdefa929097624be08b8217b3ab03b6f091dd712665a7b3bc21"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 7404,
+          "sha256": "4d6781bdd3baffdefa929097624be08b8217b3ab03b6f091dd712665a7b3bc21"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 7404,
+          "sha256": "4d6781bdd3baffdefa929097624be08b8217b3ab03b6f091dd712665a7b3bc21"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 10051,
+        "sha256": "19eda7f488885383787498cb081b3e9e6cdd37a64d8f20be1e8f035677e59bea"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 10051,
+          "sha256": "19eda7f488885383787498cb081b3e9e6cdd37a64d8f20be1e8f035677e59bea"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 10051,
+          "sha256": "19eda7f488885383787498cb081b3e9e6cdd37a64d8f20be1e8f035677e59bea"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 1825,
+        "sha256": "12f7b713dd4b203b6d5043e04600a47123a4b098e243388245849f85f9632b47"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 1825,
+          "sha256": "12f7b713dd4b203b6d5043e04600a47123a4b098e243388245849f85f9632b47"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 1825,
+          "sha256": "12f7b713dd4b203b6d5043e04600a47123a4b098e243388245849f85f9632b47"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 5594,
+        "sha256": "0d9a271a814c09b0be12effce60b59cb9d67a98fa532dca33a84f1332de362f3"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 5594,
+          "sha256": "0d9a271a814c09b0be12effce60b59cb9d67a98fa532dca33a84f1332de362f3"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 5594,
+          "sha256": "0d9a271a814c09b0be12effce60b59cb9d67a98fa532dca33a84f1332de362f3"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 2225,
+        "sha256": "0c2f6c5685c474082d41d1358f277afc04f2a1150b748979e7ac93e06e42b730"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2225,
+          "sha256": "0c2f6c5685c474082d41d1358f277afc04f2a1150b748979e7ac93e06e42b730"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2225,
+          "sha256": "0c2f6c5685c474082d41d1358f277afc04f2a1150b748979e7ac93e06e42b730"
         }
       },
       "failures": []
       "local": {
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/summary.json",
         "exists": true,
+        "bytes": 5951,
+        "sha256": "910477d2fba648605dda128d0ecd2a2c13cfa460573e350dc850014ac91c6c2b"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/summary.json",
           "exists": true,
+          "bytes": 5951,
+          "sha256": "910477d2fba648605dda128d0ecd2a2c13cfa460573e350dc850014ac91c6c2b"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/summary.json",
           "exists": true,
+          "bytes": 5951,
+          "sha256": "910477d2fba648605dda128d0ecd2a2c13cfa460573e350dc850014ac91c6c2b"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 2398,
+        "sha256": "8fc3464d2183c8e1604b499262327ef315ad0eef0f7c66464fb1229ed23e7a8d"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2398,
+          "sha256": "8fc3464d2183c8e1604b499262327ef315ad0eef0f7c66464fb1229ed23e7a8d"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2398,
+          "sha256": "8fc3464d2183c8e1604b499262327ef315ad0eef0f7c66464fb1229ed23e7a8d"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 2049,
+        "sha256": "0d3b07e5b66015b6f3cfa61e2fddcc669e2397b3659367f58eb0371b7955413d"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 2049,
+          "sha256": "0d3b07e5b66015b6f3cfa61e2fddcc669e2397b3659367f58eb0371b7955413d"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 2049,
+          "sha256": "0d3b07e5b66015b6f3cfa61e2fddcc669e2397b3659367f58eb0371b7955413d"
         }
       },
       "failures": []
       "local": {
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
         "exists": true,
+        "bytes": 1512,
+        "sha256": "12ea92277ddc22fdd29adc3c7288ca0e0ebec3be3ffdc47f1ff7c4ec075dd015"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/RUN_REPORT.md",
           "exists": true,
+          "bytes": 1512,
+          "sha256": "12ea92277ddc22fdd29adc3c7288ca0e0ebec3be3ffdc47f1ff7c4ec075dd015"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "space_results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
+        "exists": true,
+        "bytes": 12958,
+        "sha256": "0ea0cacdea44ed714fd47166a999606d76200a30453a39142554cc6bf1b12269"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
+          "exists": true,
+          "bytes": 12958,
+          "sha256": "0ea0cacdea44ed714fd47166a999606d76200a30453a39142554cc6bf1b12269"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "space_results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv",
+        "exists": true,
+        "bytes": 12133,
+        "sha256": "579c606563073db2ff09a636b294b41aad1ec8d0fa8ae19b8649d3273cf88a51"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 12133,
+          "sha256": "579c606563073db2ff09a636b294b41aad1ec8d0fa8ae19b8649d3273cf88a51"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "space_results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv",
+        "exists": true,
+        "bytes": 125863,
+        "sha256": "e8a55a2b42315e2febea992138f9ae31a5543c81219127a25f00f90cdc2440ba"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv",
+          "exists": true,
+          "bytes": 125863,
+          "sha256": "e8a55a2b42315e2febea992138f9ae31a5543c81219127a25f00f90cdc2440ba"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 21876,
+        "sha256": "af47c1007e54e06553578a2fee4c9c6a9cbd730bbef0330a32ffd846bb364f14"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 21876,
+          "sha256": "af47c1007e54e06553578a2fee4c9c6a9cbd730bbef0330a32ffd846bb364f14"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
         "exists": true,
         "bytes": 169938,
+        "sha256": "d1f1c299960ad62c684fe8b798467b49157b75f6bc89e8a10066cda71b66e305"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json",
           "exists": true,
           "bytes": 169938,
+          "sha256": "d1f1c299960ad62c684fe8b798467b49157b75f6bc89e8a10066cda71b66e305"
         }
       },
       "failures": []
       },
       "failures": []
     },
+    {
+      "name": "space_results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
+        "exists": true,
+        "bytes": 1738,
+        "sha256": "990efd574e384b785187d442858f17d6da1342ff4ab435ee569e5aa1f6faebc3"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
+          "exists": true,
+          "bytes": 1738,
+          "sha256": "990efd574e384b785187d442858f17d6da1342ff4ab435ee569e5aa1f6faebc3"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "space_results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv",
+        "exists": true,
+        "bytes": 57340,
+        "sha256": "4216a73df04338edf271aaeafe66b041675726d21909e1eb25e7face575c567f"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv",
+          "exists": true,
+          "bytes": 57340,
+          "sha256": "4216a73df04338edf271aaeafe66b041675726d21909e1eb25e7face575c567f"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "space_results/omni_finetune/model_output_task_probes_20260616/cosmos3_nano_future_window_source_window_map.jsonl",
       "status": "pass",
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 7404,
+        "sha256": "4d6781bdd3baffdefa929097624be08b8217b3ab03b6f091dd712665a7b3bc21"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 7404,
+          "sha256": "4d6781bdd3baffdefa929097624be08b8217b3ab03b6f091dd712665a7b3bc21"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 10051,
+        "sha256": "19eda7f488885383787498cb081b3e9e6cdd37a64d8f20be1e8f035677e59bea"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/long_horizon_next_action/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 10051,
+          "sha256": "19eda7f488885383787498cb081b3e9e6cdd37a64d8f20be1e8f035677e59bea"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 1825,
+        "sha256": "12f7b713dd4b203b6d5043e04600a47123a4b098e243388245849f85f9632b47"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/modality_reconstruction/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 1825,
+          "sha256": "12f7b713dd4b203b6d5043e04600a47123a4b098e243388245849f85f9632b47"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 5594,
+        "sha256": "0d9a271a814c09b0be12effce60b59cb9d67a98fa532dca33a84f1332de362f3"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 5594,
+          "sha256": "0d9a271a814c09b0be12effce60b59cb9d67a98fa532dca33a84f1332de362f3"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 2225,
+        "sha256": "0c2f6c5685c474082d41d1358f277afc04f2a1150b748979e7ac93e06e42b730"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2225,
+          "sha256": "0c2f6c5685c474082d41d1358f277afc04f2a1150b748979e7ac93e06e42b730"
         }
       },
       "failures": []
       "local": {
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/summary.json",
         "exists": true,
+        "bytes": 5951,
+        "sha256": "910477d2fba648605dda128d0ecd2a2c13cfa460573e350dc850014ac91c6c2b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/summary.json",
           "exists": true,
+          "bytes": 5951,
+          "sha256": "910477d2fba648605dda128d0ecd2a2c13cfa460573e350dc850014ac91c6c2b"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
         "exists": true,
         "bytes": 2398,
+        "sha256": "8fc3464d2183c8e1604b499262327ef315ad0eef0f7c66464fb1229ed23e7a8d"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json",
           "exists": true,
           "bytes": 2398,
+          "sha256": "8fc3464d2183c8e1604b499262327ef315ad0eef0f7c66464fb1229ed23e7a8d"
         }
       },
       "failures": []
         "path": "repo:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
         "exists": true,
         "bytes": 2049,
+        "sha256": "0d3b07e5b66015b6f3cfa61e2fddcc669e2397b3659367f58eb0371b7955413d"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json",
           "exists": true,
           "bytes": 2049,
+          "sha256": "0d3b07e5b66015b6f3cfa61e2fddcc669e2397b3659367f58eb0371b7955413d"
         }
       },
       "failures": []
       "local": {
         "path": "repo:TASK_METHOD_20_GAP_AUDIT.md",
         "exists": true,
+        "bytes": 8428,
+        "sha256": "b33c7f91565a2603b138eb4ec1c6df32d823225b16563eba279f416b0fe2a964"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:TASK_METHOD_20_GAP_AUDIT.md",
           "exists": true,
+          "bytes": 8428,
+          "sha256": "b33c7f91565a2603b138eb4ec1c6df32d823225b16563eba279f416b0fe2a964"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:TASK_METHOD_20_GAP_AUDIT.md",
           "exists": true,
+          "bytes": 8428,
+          "sha256": "b33c7f91565a2603b138eb4ec1c6df32d823225b16563eba279f416b0fe2a964"
         },
         "hf_model": {
           "path": "hf_model:TASK_METHOD_20_GAP_AUDIT.md",
           "exists": true,
+          "bytes": 8428,
+          "sha256": "b33c7f91565a2603b138eb4ec1c6df32d823225b16563eba279f416b0fe2a964"
         }
       },
       "failures": []
       "local": {
         "path": "repo:TASK_METHOD_20_RESULT_MATRIX.md",
         "exists": true,
+        "bytes": 3876,
+        "sha256": "a9ce23a452ca73b6d08219ffaac1782d5baad11f9e3662138b62cb4cd5ff206f"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:TASK_METHOD_20_RESULT_MATRIX.md",
           "exists": true,
+          "bytes": 3876,
+          "sha256": "a9ce23a452ca73b6d08219ffaac1782d5baad11f9e3662138b62cb4cd5ff206f"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:TASK_METHOD_20_RESULT_MATRIX.md",
           "exists": true,
+          "bytes": 3876,
+          "sha256": "a9ce23a452ca73b6d08219ffaac1782d5baad11f9e3662138b62cb4cd5ff206f"
         },
         "hf_model": {
           "path": "hf_model:TASK_METHOD_20_RESULT_MATRIX.md",
           "exists": true,
+          "bytes": 3876,
+          "sha256": "a9ce23a452ca73b6d08219ffaac1782d5baad11f9e3662138b62cb4cd5ff206f"
         }
       },
       "failures": []

data/public_surface_qa.json CHANGED Viewed

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Public Project Surface",
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:26:00+00:00",
   "scope": "Repo README, GitHub Pages HTML, Hugging Face Space card, artifact dataset card, and model card.",
   "checks": [
     {
@@ -18,7 +18,7 @@
         "website_integrity": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-18T21:01:47+00:00"
         },
         "rendered_site_check": {
           "exists": true,
@@ -28,27 +28,27 @@
         "task_surface_integrity": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-18T21:01:45+00:00"
         },
         "source_alignment": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-18T21:01:45+00:00"
         },
         "scale_up_status": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-18T21:01:43+00:00"
         },
         "publication_package": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-18T21:02:20+00:00"
         },
         "mirror_parity": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-18T21:03:15+00:00"
         }
       },
       "failures": {}

 {
   "title": "Ropedia Xperience-10M Public Project Surface",
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:57:13+00:00",
   "scope": "Repo README, GitHub Pages HTML, Hugging Face Space card, artifact dataset card, and model card.",
   "checks": [
     {
         "website_integrity": {
           "exists": true,
           "status": "pass",
+          "generated_at_utc": "2026-06-18T22:27:45+00:00"
         },
         "rendered_site_check": {
           "exists": true,
         "task_surface_integrity": {
           "exists": true,
           "status": "pass",
+          "generated_at_utc": "2026-06-18T22:27:40+00:00"
         },
         "source_alignment": {
           "exists": true,
           "status": "pass",
+          "generated_at_utc": "2026-06-18T22:27:40+00:00"
         },
         "scale_up_status": {
           "exists": true,
           "status": "pass",
+          "generated_at_utc": "2026-06-18T22:27:40+00:00"
         },
         "publication_package": {
           "exists": true,
           "status": "pass",
+          "generated_at_utc": "2026-06-18T22:27:55+00:00"
         },
         "mirror_parity": {
           "exists": true,
           "status": "pass",
+          "generated_at_utc": "2026-06-18T22:36:57+00:00"
         }
       },
       "failures": {}

data/quality_gates.json CHANGED Viewed

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Release Checks",
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:26:00+00:00",
   "rule": "A release is current when the automated reports pass and the live GitHub/Hugging Face mirrors are verified after publishing.",
   "automated_gates": [
     {

 {
   "title": "Ropedia Xperience-10M Release Checks",
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:57:13+00:00",
   "rule": "A release is current when the automated reports pass and the live GitHub/Hugging Face mirrors are verified after publishing.",
   "automated_gates": [
     {

data/single_episode_task_model_radar.json CHANGED Viewed

@@ -1,7 +1,7 @@
 {
   "title": "Single-Episode 20-Task Radar",
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:25:59+00:00",
   "description": "Minimal and Neural MLP baselines on the one public sample episode, both scored on all 20 task contracts.",
   "task_count": 20,
   "method_count": 2,

 {
   "title": "Single-Episode 20-Task Radar",
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:55:01+00:00",
   "description": "Minimal and Neural MLP baselines on the one public sample episode, both scored on all 20 task contracts.",
   "task_count": 20,
   "method_count": 2,

data/source_alignment_audit.json CHANGED Viewed

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Source Alignment Note",
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:27:40+00:00",
   "alignment_json": "docs/data/xperience10m_dataset_card_alignment.json",
   "alignment_summary": {
     "full_dataset_repo": "ropedia-ai/xperience-10m",

 {
   "title": "Ropedia Xperience-10M Source Alignment Note",
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:57:59+00:00",
   "alignment_json": "docs/data/xperience10m_dataset_card_alignment.json",
   "alignment_summary": {
     "full_dataset_repo": "ropedia-ai/xperience-10m",

data/task_method_20_gap_audit.json CHANGED Viewed

@@ -1,10 +1,10 @@
 {
-  "generated_at_utc": "2026-06-18T22:25:59+00:00",
   "immediate_actions": [
     {
       "artifact": "docs/data/task_method_20_gap_audit.json",
       "id": "gap_audit",
-      "purpose": "Keep the 29 scoreless cells visible and reproducible."
     },
     {
       "artifact": "scripts/omni/score_model_output_probes.py",
@@ -24,11 +24,11 @@
       "proxy_scored_task_count": 0,
       "result_record_count": 20,
       "scope": "128 selected episodes, held-out test",
-      "scored_task_count": 10,
-      "scoreless_task_count": 10,
       "status_counts": {
-        "not_evaluated_in_verified_package": 10,
-        "scored": 10
       }
     },
     "cosmos3_super_reasoner": {
@@ -37,11 +37,11 @@
       "proxy_scored_task_count": 0,
       "result_record_count": 20,
       "scope": "128 selected episodes, held-out test",
-      "scored_task_count": 9,
-      "scoreless_task_count": 11,
       "status_counts": {
-        "not_evaluated_in_verified_package": 11,
-        "scored": 9
       }
     },
     "metadata128_neural_mlp": {
@@ -135,14 +135,14 @@
     }
   },
   "missing_by_method": {
-    "cosmos3_nano_future_window": 10,
-    "cosmos3_super_reasoner": 11,
     "metadata128_neural_mlp": 2,
     "metadata128_simple": 2,
     "qwen3_omni_v6_lora": 4
   },
   "missing_by_status": {
-    "not_evaluated_in_verified_package": 25,
     "not_supported_by_metadata_only_package": 2,
     "unsupported_without_required_target": 2
   },
@@ -159,8 +159,7 @@
       "cosmos3_nano_future_window"
     ],
     "08 Language Grounding": [
-      "cosmos3_nano_future_window",
-      "cosmos3_super_reasoner"
     ],
     "09 Cross-Modal Retrieval": [
       "cosmos3_super_reasoner"
@@ -187,9 +186,6 @@
       "metadata128_simple",
       "qwen3_omni_v6_lora"
     ],
-    "16 Action-Object Relation Prediction": [
-      "cosmos3_nano_future_window"
-    ],
     "17 Future Object-Set Forecasting": [
       "cosmos3_super_reasoner"
     ],
@@ -271,19 +267,6 @@
       "task_label": "Object Relevance Prediction",
       "task_number": 7
     },
-    {
-      "method": "Cosmos3-Super Reasoner",
-      "metric_key": "mrr",
-      "reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
-      "recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
-      "scope": "multi_episode_128_partial_model_overlay",
-      "series_id": "cosmos3_super_reasoner",
-      "status": "not_evaluated_in_verified_package",
-      "status_label": "not evaluated",
-      "task_id": "caption_grounding",
-      "task_label": "Language Grounding",
-      "task_number": 8
-    },
     {
       "method": "Cosmos3-Nano Future Window",
       "metric_key": "mrr",
@@ -466,19 +449,6 @@
       "task_label": "Interaction Text Prediction",
       "task_number": 15
     },
-    {
-      "method": "Cosmos3-Nano Future Window",
-      "metric_key": "macro_f1",
-      "reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
-      "recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
-      "scope": "multi_episode_128_partial_model_overlay",
-      "series_id": "cosmos3_nano_future_window",
-      "status": "not_evaluated_in_verified_package",
-      "status_label": "not evaluated",
-      "task_id": "action_object_relation",
-      "task_label": "Action-Object Relation Prediction",
-      "task_number": 16
-    },
     {
       "method": "Cosmos3-Super Reasoner",
       "metric_key": "micro_f1",
@@ -630,8 +600,8 @@
     "method_count": 9,
     "method_task_record_count": 180,
     "proxy_scored_method_task_count": 4,
-    "scored_method_task_count": 151,
-    "scoreless_method_task_count": 29,
     "task_count": 20
   },
   "source_matrix": "docs/data/task_method_20_result_matrix.json",

 {
+  "generated_at_utc": "2026-06-18T22:55:35+00:00",
   "immediate_actions": [
     {
       "artifact": "docs/data/task_method_20_gap_audit.json",
       "id": "gap_audit",
+      "purpose": "Keep the 27 scoreless cells visible and reproducible."
     },
     {
       "artifact": "scripts/omni/score_model_output_probes.py",
       "proxy_scored_task_count": 0,
       "result_record_count": 20,
       "scope": "128 selected episodes, held-out test",
+      "scored_task_count": 11,
+      "scoreless_task_count": 9,
       "status_counts": {
+        "not_evaluated_in_verified_package": 9,
+        "scored": 11
       }
     },
     "cosmos3_super_reasoner": {
       "proxy_scored_task_count": 0,
       "result_record_count": 20,
       "scope": "128 selected episodes, held-out test",
+      "scored_task_count": 10,
+      "scoreless_task_count": 10,
       "status_counts": {
+        "not_evaluated_in_verified_package": 10,
+        "scored": 10
       }
     },
     "metadata128_neural_mlp": {
     }
   },
   "missing_by_method": {
+    "cosmos3_nano_future_window": 9,
+    "cosmos3_super_reasoner": 10,
     "metadata128_neural_mlp": 2,
     "metadata128_simple": 2,
     "qwen3_omni_v6_lora": 4
   },
   "missing_by_status": {
+    "not_evaluated_in_verified_package": 23,
     "not_supported_by_metadata_only_package": 2,
     "unsupported_without_required_target": 2
   },
       "cosmos3_nano_future_window"
     ],
     "08 Language Grounding": [
+      "cosmos3_nano_future_window"
     ],
     "09 Cross-Modal Retrieval": [
       "cosmos3_super_reasoner"
       "metadata128_simple",
       "qwen3_omni_v6_lora"
     ],
     "17 Future Object-Set Forecasting": [
       "cosmos3_super_reasoner"
     ],
       "task_label": "Object Relevance Prediction",
       "task_number": 7
     },
     {
       "method": "Cosmos3-Nano Future Window",
       "metric_key": "mrr",
       "task_label": "Interaction Text Prediction",
       "task_number": 15
     },
     {
       "method": "Cosmos3-Super Reasoner",
       "metric_key": "micro_f1",
     "method_count": 9,
     "method_task_record_count": 180,
     "proxy_scored_method_task_count": 4,
+    "scored_method_task_count": 153,
+    "scoreless_method_task_count": 27,
     "task_count": 20
   },
   "source_matrix": "docs/data/task_method_20_result_matrix.json",

data/task_method_20_result_matrix.json CHANGED Viewed

@@ -1,11 +1,11 @@
 {
   "title": "Task Method 20-Result Matrix",
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:25:59+00:00",
   "task_count": 20,
   "method_count": 9,
   "method_task_record_count": 180,
-  "scored_method_task_count": 151,
   "series": [
     {
       "id": "minimal",
@@ -181,20 +181,20 @@
       "kind": "partial_128_episode_foundation_model_overlay",
       "scope": "128 selected episodes, held-out test",
       "stroke_dasharray": "4 7",
-      "method_detail": "Verified Cosmos3-Super base-weight Reasoner JSON-task evaluation, plus task 16 and a derived task-20 action-boundary timing probe scored from existing verified JSON.",
       "plotted_as": "colored point overlay",
       "result_record_count": 20,
-      "scored_task_count": 9,
-      "covered_task_count": 9,
       "proxy_scored_task_count": 0,
-      "scoreless_task_count": 11,
       "unsupported_task_count": 0,
-      "not_evaluated_task_count": 11,
       "status_counts": {
-        "not_evaluated_in_verified_package": 11,
-        "scored": 9
       },
-      "coverage_fraction": 0.45,
       "result_record_fraction": 1.0
     },
     {
@@ -205,20 +205,20 @@
       "kind": "partial_128_episode_world_model_overlay",
       "scope": "128 selected episodes, held-out test",
       "stroke_dasharray": "2 7",
-      "method_detail": "Verified Cosmos3-Nano future-window compatibility metrics, plus tasks 10/13/14/17 and a derived task-20 boundary timing probe scored from existing held-out future-window artifacts.",
       "plotted_as": "colored point overlay",
       "result_record_count": 20,
-      "scored_task_count": 10,
-      "covered_task_count": 10,
       "proxy_scored_task_count": 0,
-      "scoreless_task_count": 10,
       "unsupported_task_count": 0,
-      "not_evaluated_task_count": 10,
       "status_counts": {
-        "not_evaluated_in_verified_package": 10,
-        "scored": 10
       },
-      "coverage_fraction": 0.5,
       "result_record_fraction": 1.0
     }
   ],
@@ -1489,17 +1489,17 @@
       "task_label": "Language Grounding",
       "series_id": "cosmos3_super_reasoner",
       "method": "Cosmos3-Super Reasoner",
-      "status": "not_evaluated_in_verified_package",
-      "status_label": "not evaluated",
-      "scored": false,
       "proxy_scored": false,
-      "raw": null,
-      "raw_text": "n/a",
-      "normalized_score": null,
-      "metric_key": "mrr",
-      "source": null,
       "scope": "multi_episode_128_partial_model_overlay",
-      "reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score"
     },
     {
       "task_number": 8,
@@ -2803,17 +2803,17 @@
       "task_label": "Action-Object Relation Prediction",
       "series_id": "cosmos3_nano_future_window",
       "method": "Cosmos3-Nano Future Window",
-      "status": "not_evaluated_in_verified_package",
-      "status_label": "not evaluated",
-      "scored": false,
       "proxy_scored": false,
-      "raw": null,
-      "raw_text": "n/a",
-      "normalized_score": null,
-      "metric_key": "macro_f1",
-      "source": null,
       "scope": "multi_episode_128_partial_model_overlay",
-      "reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score"
     },
     {
       "task_number": 17,

 {
   "title": "Task Method 20-Result Matrix",
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:55:01+00:00",
   "task_count": 20,
   "method_count": 9,
   "method_task_record_count": 180,
+  "scored_method_task_count": 153,
   "series": [
     {
       "id": "minimal",
       "kind": "partial_128_episode_foundation_model_overlay",
       "scope": "128 selected episodes, held-out test",
       "stroke_dasharray": "4 7",
+      "method_detail": "Verified Cosmos3-Super base-weight Reasoner JSON-task evaluation, plus task 8/16 and a derived task-20 action-boundary timing probe scored from existing verified JSON.",
       "plotted_as": "colored point overlay",
       "result_record_count": 20,
+      "scored_task_count": 10,
+      "covered_task_count": 10,
       "proxy_scored_task_count": 0,
+      "scoreless_task_count": 10,
       "unsupported_task_count": 0,
+      "not_evaluated_task_count": 10,
       "status_counts": {
+        "not_evaluated_in_verified_package": 10,
+        "scored": 10
       },
+      "coverage_fraction": 0.5,
       "result_record_fraction": 1.0
     },
     {
       "kind": "partial_128_episode_world_model_overlay",
       "scope": "128 selected episodes, held-out test",
       "stroke_dasharray": "2 7",
+      "method_detail": "Verified Cosmos3-Nano future-window compatibility metrics, plus tasks 10/13/14/16/17 and a derived task-20 boundary timing probe scored from existing held-out future-window artifacts.",
       "plotted_as": "colored point overlay",
       "result_record_count": 20,
+      "scored_task_count": 11,
+      "covered_task_count": 11,
       "proxy_scored_task_count": 0,
+      "scoreless_task_count": 9,
       "unsupported_task_count": 0,
+      "not_evaluated_task_count": 9,
       "status_counts": {
+        "not_evaluated_in_verified_package": 9,
+        "scored": 11
       },
+      "coverage_fraction": 0.55,
       "result_record_fraction": 1.0
     }
   ],
       "task_label": "Language Grounding",
       "series_id": "cosmos3_super_reasoner",
       "method": "Cosmos3-Super Reasoner",
+      "status": "scored",
+      "status_label": "scored",
+      "scored": true,
       "proxy_scored": false,
+      "raw": 0.30639899644580487,
+      "raw_text": "0.3064",
+      "normalized_score": 0.30639899644580487,
+      "metric_key": "caption_grounding_iou",
+      "source": "results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
       "scope": "multi_episode_128_partial_model_overlay",
+      "reason": null
     },
     {
       "task_number": 8,
       "task_label": "Action-Object Relation Prediction",
       "series_id": "cosmos3_nano_future_window",
       "method": "Cosmos3-Nano Future Window",
+      "status": "scored",
+      "status_label": "scored",
+      "scored": true,
       "proxy_scored": false,
+      "raw": 0.002794157670325683,
+      "raw_text": "0.0028",
+      "normalized_score": 0.002794157670325683,
+      "metric_key": "action_object_relation_macro_f1",
+      "source": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
       "scope": "multi_episode_128_partial_model_overlay",
+      "reason": null
     },
     {
       "task_number": 17,

data/task_surface_integrity.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:27:40+00:00",
   "summary": {
     "task_count": 12,
     "expected_task_count": 12,

 {
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:57:59+00:00",
   "summary": {
     "task_count": 12,
     "expected_task_count": 12,

data/website_integrity.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-18T22:27:45+00:00",
   "docs_root": "docs",
   "site_base": "/ropedia-xperience-10m-task-suite/",
   "summary": {
@@ -301,7 +301,7 @@
     },
     {
       "path": "data/artifact_index.json",
-      "bytes": 116643,
       "top_level_type": "dict"
     },
     {
@@ -316,7 +316,7 @@
     },
     {
       "path": "data/episode128_task_model_radar.json",
-      "bytes": 185391,
       "top_level_type": "dict"
     },
     {
@@ -351,7 +351,7 @@
     },
     {
       "path": "data/mirror_parity.json",
-      "bytes": 938745,
       "top_level_type": "dict"
     },
     {
@@ -486,12 +486,12 @@
     },
     {
       "path": "data/task_method_20_gap_audit.json",
-      "bytes": 29788,
       "top_level_type": "dict"
     },
     {
       "path": "data/task_method_20_result_matrix.json",
-      "bytes": 128842,
       "top_level_type": "dict"
     },
     {
@@ -526,7 +526,7 @@
     },
     {
       "path": "data/unified_task_model_radar.json",
-      "bytes": 229244,
       "top_level_type": "dict"
     },
     {
@@ -571,7 +571,7 @@
     {
       "path": "assets/charts/episode128_task_model_radar.svg",
       "exists": true,
-      "bytes": 48704,
       "format": "SVG",
       "has_viewbox": true
     },
@@ -641,7 +641,7 @@
     {
       "path": "assets/charts/unified_task_model_radar.svg",
       "exists": true,
-      "bytes": 54717,
       "format": "SVG",
       "has_viewbox": true
     },

 {
   "status": "pass",
+  "generated_at_utc": "2026-06-18T22:58:01+00:00",
   "docs_root": "docs",
   "site_base": "/ropedia-xperience-10m-task-suite/",
   "summary": {
     },
     {
       "path": "data/artifact_index.json",
+      "bytes": 116644,
       "top_level_type": "dict"
     },
     {
     },
     {
       "path": "data/episode128_task_model_radar.json",
+      "bytes": 185314,
       "top_level_type": "dict"
     },
     {
     },
     {
       "path": "data/mirror_parity.json",
+      "bytes": 1015184,
       "top_level_type": "dict"
     },
     {
     },
     {
       "path": "data/task_method_20_gap_audit.json",
+      "bytes": 28259,
       "top_level_type": "dict"
     },
     {
       "path": "data/task_method_20_result_matrix.json",
+      "bytes": 128805,
       "top_level_type": "dict"
     },
     {
     },
     {
       "path": "data/unified_task_model_radar.json",
+      "bytes": 229168,
       "top_level_type": "dict"
     },
     {
     {
       "path": "assets/charts/episode128_task_model_radar.svg",
       "exists": true,
+      "bytes": 48923,
       "format": "SVG",
       "has_viewbox": true
     },
     {
       "path": "assets/charts/unified_task_model_radar.svg",
       "exists": true,
+      "bytes": 54936,
       "format": "SVG",
       "has_viewbox": true
     },

results/omni_finetune/model_output_probe_readiness/RUN_REPORT.md CHANGED Viewed

@@ -1,6 +1,6 @@
 # Model Output Probe Readiness
-Generated: `2026-06-18T18:10:07+00:00`
 This report checks whether verified model branches have the prediction files
 needed to extend them to every 20-task contract. It is readiness evidence only;
@@ -8,6 +8,6 @@ it does not assign new task scores.
 | Method | ID | Matrix scores | Status | Split files | Next step |
 | --- | --- | --- | --- | --- | --- |
-| Cosmos3-Nano Future Window | cosmos3_nano_future_window | 7/20 | missing_required_model_outputs | train: missing; validation: missing; test: missing | Collect or generate train, validation, and test prediction JSONL files first. |
-| Cosmos3-Super Reasoner | cosmos3_super_reasoner | 8/20 | missing_required_model_outputs | train: missing; validation: missing; test: present | Collect or generate train, validation, and test prediction JSONL files first. |
-| Qwen3-Omni v6 LoRA | qwen3_omni_v6_lora | 15/20 | missing_required_model_outputs | train: missing; validation: missing; test: present | Collect or generate train, validation, and test prediction JSONL files first. |

 # Model Output Probe Readiness
+Generated: `2026-06-18T22:44:42+00:00`
 This report checks whether verified model branches have the prediction files
 needed to extend them to every 20-task contract. It is readiness evidence only;
 | Method | ID | Matrix scores | Status | Split files | Next step |
 | --- | --- | --- | --- | --- | --- |
+| Cosmos3-Nano Future Window | cosmos3_nano_future_window | 10/20 | missing_required_model_outputs | train: missing; validation: missing; test: missing | Collect or generate train, validation, and test prediction JSONL files first. |
+| Cosmos3-Super Reasoner | cosmos3_super_reasoner | 9/20 | missing_required_model_outputs | train: missing; validation: missing; test: present | Collect or generate train, validation, and test prediction JSONL files first. |
+| Qwen3-Omni v6 LoRA | qwen3_omni_v6_lora | 16/20 | missing_required_model_outputs | train: missing; validation: missing; test: present | Collect or generate train, validation, and test prediction JSONL files first. |

results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json CHANGED Viewed

@@ -1,10 +1,10 @@
 {
-  "generated_at_utc": "2026-06-18T18:10:07+00:00",
   "methods": {
     "cosmos3_nano_future_window": {
       "label": "Cosmos3-Nano Future Window",
-      "matrix_scored_task_count": 7,
-      "matrix_scoreless_task_count": 13,
       "next_step": "Collect or generate train, validation, and test prediction JSONL files first.",
       "ready_for_all_task_probe": false,
       "required_splits": [
@@ -19,13 +19,10 @@
         "caption_grounding",
         "temporal_order",
         "misalignment_detection",
-        "next_subtask_forecast",
         "interaction_text_prediction",
         "action_object_relation",
-        "object_set_forecast",
         "imu_to_hand_pose",
-        "camera_view_sync_retrieval",
-        "time_to_transition"
       ],
       "split_status": {
         "test": {
@@ -54,8 +51,8 @@
     },
     "cosmos3_super_reasoner": {
       "label": "Cosmos3-Super Reasoner",
-      "matrix_scored_task_count": 8,
-      "matrix_scoreless_task_count": 12,
       "next_step": "Collect or generate train, validation, and test prediction JSONL files first.",
       "ready_for_all_task_probe": false,
       "required_splits": [
@@ -70,7 +67,6 @@
         "modality_reconstruction",
         "temporal_order",
         "misalignment_detection",
-        "long_horizon_next_action",
         "next_subtask_forecast",
         "interaction_text_prediction",
         "object_set_forecast",
@@ -103,8 +99,8 @@
     },
     "qwen3_omni_v6_lora": {
       "label": "Qwen3-Omni v6 LoRA",
-      "matrix_scored_task_count": 15,
-      "matrix_scoreless_task_count": 5,
       "next_step": "Collect or generate train, validation, and test prediction JSONL files first.",
       "ready_for_all_task_probe": false,
       "required_splits": [
@@ -116,8 +112,7 @@
         "hand_trajectory_forecast",
         "modality_reconstruction",
         "interaction_text_prediction",
-        "imu_to_hand_pose",
-        "camera_view_sync_retrieval"
       ],
       "split_status": {
         "test": {

 {
+  "generated_at_utc": "2026-06-18T22:44:42+00:00",
   "methods": {
     "cosmos3_nano_future_window": {
       "label": "Cosmos3-Nano Future Window",
+      "matrix_scored_task_count": 10,
+      "matrix_scoreless_task_count": 10,
       "next_step": "Collect or generate train, validation, and test prediction JSONL files first.",
       "ready_for_all_task_probe": false,
       "required_splits": [
         "caption_grounding",
         "temporal_order",
         "misalignment_detection",
         "interaction_text_prediction",
         "action_object_relation",
         "imu_to_hand_pose",
+        "camera_view_sync_retrieval"
       ],
       "split_status": {
         "test": {
     },
     "cosmos3_super_reasoner": {
       "label": "Cosmos3-Super Reasoner",
+      "matrix_scored_task_count": 9,
+      "matrix_scoreless_task_count": 11,
       "next_step": "Collect or generate train, validation, and test prediction JSONL files first.",
       "ready_for_all_task_probe": false,
       "required_splits": [
         "modality_reconstruction",
         "temporal_order",
         "misalignment_detection",
         "next_subtask_forecast",
         "interaction_text_prediction",
         "object_set_forecast",
     },
     "qwen3_omni_v6_lora": {
       "label": "Qwen3-Omni v6 LoRA",
+      "matrix_scored_task_count": 16,
+      "matrix_scoreless_task_count": 4,
       "next_step": "Collect or generate train, validation, and test prediction JSONL files first.",
       "ready_for_all_task_probe": false,
       "required_splits": [
         "hand_trajectory_forecast",
         "modality_reconstruction",
         "interaction_text_prediction",
+        "imu_to_hand_pose"
       ],
       "split_status": {
         "test": {

results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json ADDED Viewed

	@@ -0,0 +1,187 @@

+{
+  "accuracy": 0.013297872340425532,
+  "action_object_relation_accuracy": 0.013297872340425532,
+  "action_object_relation_macro_f1": 0.002794157670325683,
+  "artifact_files": {
+    "metrics_json": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json",
+    "per_class_metrics_csv": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv",
+    "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv"
+  },
+  "generated_at_utc": "2026-06-18T22:52:18+00:00",
+  "known_limitation": "This is a retrieval-derived future relation score, not a separately prompted action-object generation head.",
+  "labels": [
+    "Adjust canned food on shelf :: canned food | cardboard box | store shelf",
+    "Adjust item on shelf :: shelf | stationery package",
+    "Adjust lantern shape :: red pleated paper lantern",
+    "Adjust paper :: cardboard square | paper | pen | star beads",
+    "Adjust pot position :: pot",
+    "Adjust puzzle piece :: jigsaw puzzle | puzzle box | puzzle pieces",
+    "Align edges of paper lantern :: hands | red paper lantern",
+    "Align paper lantern edges :: hands | red paper lantern",
+    "Apply adhesive tape to lantern :: adhesive tape | red paper lantern",
+    "Approach boxes :: cardboard boxes | colleague | shelving unit",
+    "Approaching and pressing the door switch :: glass door | switch | wall",
+    "Approaching the table :: beads | chairs | table",
+    "Arrange buttons :: red buttons | smartphone | table",
+    "Arrange buttons in a line :: buttons | phone | table",
+    "Arrange star beads for counting :: paper | pen | star beads",
+    "Attempt to fit puzzle piece :: puzzle board | puzzle piece",
+    "Attempt to fit puzzle piece :: puzzle piece",
+    "Bend and manipulate plastic strip :: plastic strip | stationary box | table | water bottle",
+    "Browse smartphone screen :: paper strips | smartphone | star ornaments",
+    "Closing the door :: door | dustpan",
+    "Counting and organizing beads :: cardboard squares | paper | pen | star-shaped beads",
+    "Counting star beads :: paper | pen | star beads",
+    "Cut along the marked line :: cardboard | hand | utility knife",
+    "Cut cardboard piece :: cardboard strip | scissors",
+    "Entering the VR training room :: doorway | person | table",
+    "Expand paper lantern :: adhesive tape roll | cardboard box | red paper lantern",
+    "Extract wire hangers from box :: cardboard box | wire hangers",
+    "Fold paper lantern :: red paper honeycomb lantern",
+    "Gesturing :: cardboard pieces | marker | pencil case | ruler | scissors",
+    "Grasp cleaning bottle :: bowls | chopping board | cleaning fluid bottle",
+    "Grasp lantern :: red paper honeycomb lantern",
+    "Grasping cleaning cloth :: cleaning cloth | countertop",
+    "Greeting/acknowledging participants :: person | table | vr headset",
+    "Handle paper lantern component :: cardboard box | paper lantern component | plastic bag | red paper lantern",
+    "Hold and bend plastic strip :: bottle | purple plastic strip | stationery box",
+    "Hold and manipulate paper strip :: beads | mobile phone | power bank | yellow paper strip",
+    "Hold and manipulate paper strip :: beads | paper strip | power bank | smartphone",
+    "Hold beads :: paper | pen | power bank | smartphone | star-shaped beads",
+    "Hold container lid :: dustpan | red plastic container lid",
+    "Hold earbud case :: box of products | earbud case | shelf",
+    "Hold smartphone :: quilling paper strips | small blue beads | smartphone",
+    "Identify next cardboard piece :: cardboard pieces | marker",
+    "Inspect shelf condition :: colleague | shelf",
+    "Interact with smartphone :: beads | smartphone | yellow paper strip",
+    "Interact with smartphone :: cans | cereal box | shelf | smartphone",
+    "Interact with smartphone :: cans | shelf | smartphone",
+    "Lift pot lid :: pot | pot lid",
+    "Manipulate adhesive strip :: adhesive strip | paper sheets | puzzle box | smartphone | water bottle",
+    "Manipulate bead :: bead piles | container | paper strips | yellow bead",
+    "Manipulate bead :: bead | bead piles | pen holder | table",
+    "Manipulate bead :: bead | bead piles | table",
+    "Manipulate beads :: blue beads | power bank | smartphone | yellow beads",
+    "Manipulate craft paper strips :: paper strips | scissors | smartphone",
+    "Manipulate craft piece :: craft pieces | scissors | smartphone",
+    "Manipulate material :: blue crafting material | hand",
+    "Manipulate paper edge :: hands | paper cone | puzzle box | smartphone | water bottle",
+    "Manipulate paper strip :: container with tools | table | yellow paper strip",
+    "Manipulate paper strip :: craft materials | purple paper strip | scissors | smartphone",
+    "Manipulate paper strip :: paper cone | paper strip | puzzle box | smartphone | water bottle",
+    "Manipulate paper strip :: purple paper strip | quilling paper pile | smartphone",
+    "Manipulate plastic strip :: purple plastic strip | storage bin | water bottle",
+    "Manipulate plastic strips :: desk | plastic strips | water bottle",
+    "Manipulate puzzle pieces :: puzzle box | puzzle mat | puzzle pieces",
+    "Manipulate yellow strip :: beads | cell phone | pen | yellow strip",
+    "Manipulating paper strips :: beads | blue paper strip | yellow paper strip",
+    "Marking cardboard piece :: cardboard pieces | marker | ruler | scissors",
+    "Move dustpan to side :: dustpan",
+    "Move marker and adjust hand :: cardboard pieces | marker",
+    "Move phone :: cardboard | smartphone | utility knife",
+    "Move pot :: faucet | pot | sink",
+    "Move smartphone :: craft beads | paper strips | smartphone",
+    "Observe and pause :: cardboard pieces | marker | pencil case | ruler | scissors",
+    "Open earbud case :: earbud case",
+    "Open folded paper lantern :: cardboard box | red pleated paper lantern",
+    "Open paper lantern component :: paper lantern component",
+    "Open stove pot lid :: blue bowl | cloth | faucet | red bowl | sink | soap dispenser | white bowl",
+    "Operate smartphone :: craft paper strips | folded paper fan | paper scraps | scissors | smartphone | table",
+    "Operate smartphone :: craft paper strips | folded paper fan | smartphone",
+    "Operate smartphone :: paper scraps | scissors | smartphone | table",
+    "Organize cardboard pieces :: cardboard pieces | marker | pencil case | ruler | scissors",
+    "Pick up button :: buttons | table",
+    "Pick up items from the shopping bag :: cardboard boxes | red shopping bag | retail shelf",
+    "Pick up new cardboard piece :: cardboard piece | marker",
+    "Pick up packaged paper lantern component :: cardboard box | packaged paper lantern component",
+    "Pick up smartphone :: cans | shelf | smartphone",
+    "Pick up star bead :: cardboard squares | mobile phone | paper | pen | power bank | star beads",
+    "Pick up utility knife :: cardboard | utility knife",
+    "Picking up bottle :: bottle",
+    "Picking up crafting material :: beads | smartphone | table | yellow strip",
+    "Place and count bead :: paper | pen | star beads",
+    "Place another canned food on shelf :: box | canned food",
+    "Place button :: buttons | hand | smartphone | table",
+    "Place can on shelf :: canned food | retail shelf",
+    "Place canned food on shelf :: canned food | store shelf",
+    "Place cloth on floor :: cloth",
+    "Place item on shelf :: stationery package",
+    "Place items on the shelf :: packaged items | retail shelf",
+    "Place material :: bead design | blue crafting material | hand",
+    "Place phone down :: beads | pencil holder | smartphone | yellow paper strip",
+    "Place piece into puzzle :: jigsaw puzzle | puzzle piece",
+    "Place puzzle piece :: puzzle board | puzzle piece",
+    "Place smartphone down :: power bank | quilling paper strips | small paper stars | smartphone",
+    "Place smartphone down :: quilling paper strips | small paper stars | smartphone",
+    "Place smartphone on stand :: saucepan | smartphone | smartphone stand | yellow jacket",
+    "Place towel :: pot | towel",
+    "Placing paper strip :: adhesive strip | paper cone | phone",
+    "Put down smartphone :: paper quilling strips | power bank | smartphone | star beads",
+    "Put down smartphone :: paper quilling strips | smartphone | star beads",
+    "Put down smartphone :: power bank | smartphone",
+    "Reach for another item :: item | packaged item | shelf",
+    "Reach for cleaning supplies :: bowls | chopping board | cleaning fluid bottle",
+    "Reach for next canned food :: box | canned food",
+    "Reach for next canned food :: box | canned food | retail shelf",
+    "Reach for next item :: packaged item | shelf",
+    "Reach for puzzle piece :: puzzle board | puzzle box | puzzle pieces",
+    "Reach for wire hangers :: cardboard box | wire hangers",
+    "Reach into box :: box of cans",
+    "Release cardboard piece and gesture :: cardboard piece | cardboard piles | marker | pouch | ruler | scissors",
+    "Release hook :: display hook",
+    "Release lantern :: red paper lantern",
+    "Remove paper lantern part from packaging :: paper lantern component | plastic packaging",
+    "Remove paper lantern part from packaging :: paper lantern | red hand fan",
+    "Remove plastic packaging :: packaging | paper lantern component",
+    "Reposition hand :: cardboard | utility knife",
+    "Resume observation :: cardboard pieces | marker | pencil case | ruler | scissors",
+    "Retrieve canned food from box :: box | canned food",
+    "Retrieve next canned food item :: canned food | cardboard box",
+    "Retrieving more beads :: paper | pen | star beads",
+    "Rinse cloth in sink :: cloth | sink | water faucet",
+    "Scroll smartphone screen :: paper stars | paper strips | smartphone",
+    "Search for puzzle piece :: jigsaw puzzle | puzzle pieces",
+    "Secure paper edges with adhesive :: adhesive strip | paper cone pieces | puzzle box | smartphone | water bottle",
+    "Secure paper edges with adhesive :: adhesive strip | paper cone | paper cone pieces | puzzle box | smartphone | water bottle",
+    "Securing paper structure :: paper cone | paper strip | puzzle box | smartphone | water bottle",
+    "Sort and adjust button line :: buttons | smartphone | table",
+    "Sort and arrange buttons :: buttons | smartphones | soda can | table",
+    "Sort and arrange buttons :: buttons | smartphones | table",
+    "Sort and count beads :: cell phone | paper | pen | star beads",
+    "Sort and count beads :: paper | pen | star beads",
+    "Sort beads and write count :: paper | pen | power bank | smartphone | star-shaped beads",
+    "Sort button :: buttons | cell phone",
+    "Sort craft items :: star-shaped craft items | table",
+    "Sort puzzle pieces :: jigsaw puzzle pieces | table",
+    "Sort star-shaped beads :: marker | mobile phone | paper | power bank | star-shaped beads",
+    "Stir contents :: cooking utensil | pot",
+    "Use smartphone :: buttons | chair | smartphone | table",
+    "Use smartphone :: buttons | smartphone | table",
+    "Use smartphone :: charging cable | paper strips | power bank | smartphone | star-shaped paper crafts",
+    "Walk towards shelves :: cardboard boxes | red bin | shelving unit",
+    "Walking towards door :: door | dustpan",
+    "Write count on paper :: marker | mobile phone | paper | power bank | star-shaped beads"
+  ],
+  "macro_f1": 0.002794157670325683,
+  "metric_key": "action_object_relation_macro_f1",
+  "missing_pred_target_count": 0,
+  "missing_true_target_count": 0,
+  "model_id": "cosmos3_nano_future_window",
+  "model_label": "Cosmos3-Nano Future Window",
+  "normalization_policy": "Action text is whitespace-normalized. Objects are casefolded, deduplicated, sorted, and joined into a canonical relation label before macro-F1 scoring.",
+  "num_samples": 376,
+  "primary_metric": "action_object_relation_macro_f1",
+  "primary_score": 0.002794157670325683,
+  "scope": "held_out_test_existing_future_window_retrieval_probe",
+  "score_policy": "Derived from existing verified held-out Cosmos3-Nano future-window predictions. The true future record and retrieved future record are joined to the public-safe target map, then scored on the action plus object-set relation.",
+  "scored_rows": 376,
+  "source_prediction_jsonl": "results/omni_finetune/verified_public/xperience10m_cosmos3_nano_128ep_future_window_h5_compat_adapter_eval_test_full/eval/future_predictions.jsonl",
+  "source_target_map_jsonl": "results/omni_finetune/model_output_task_probes_20260616/cosmos3_nano_future_window_target_map.jsonl",
+  "status": "pass",
+  "target_map_rows": 448,
+  "task_id": "action_object_relation",
+  "task_label": "Action-Object Relation",
+  "task_number": 16,
+  "title": "Cosmos3-Nano Future Window Future Action-Object Relation Probe",
+  "total_prediction_rows": 378
+}

results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/per_class_metrics.csv ADDED Viewed

	@@ -0,0 +1,152 @@

+class_name,support,predicted,precision,recall,f1
+Adjust canned food on shelf :: canned food | cardboard box | store shelf,2,0,0.0,0.0,0.0
+Adjust item on shelf :: shelf | stationery package,4,1,0.0,0.0,0.0
+Adjust lantern shape :: red pleated paper lantern,2,0,0.0,0.0,0.0
+Adjust paper :: cardboard square | paper | pen | star beads,3,0,0.0,0.0,0.0
+Adjust pot position :: pot,3,0,0.0,0.0,0.0
+Adjust puzzle piece :: jigsaw puzzle | puzzle box | puzzle pieces,3,0,0.0,0.0,0.0
+Align edges of paper lantern :: hands | red paper lantern,2,0,0.0,0.0,0.0
+Align paper lantern edges :: hands | red paper lantern,2,0,0.0,0.0,0.0
+Apply adhesive tape to lantern :: adhesive tape | red paper lantern,3,0,0.0,0.0,0.0
+Approach boxes :: cardboard boxes | colleague | shelving unit,3,0,0.0,0.0,0.0
+Approaching and pressing the door switch :: glass door | switch | wall,6,0,0.0,0.0,0.0
+Approaching the table :: beads | chairs | table,2,0,0.0,0.0,0.0
+Arrange buttons :: red buttons | smartphone | table,2,0,0.0,0.0,0.0
+Arrange buttons in a line :: buttons | phone | table,7,0,0.0,0.0,0.0
+Arrange star beads for counting :: paper | pen | star beads,4,0,0.0,0.0,0.0
+Attempt to fit puzzle piece :: puzzle board | puzzle piece,3,0,0.0,0.0,0.0
+Attempt to fit puzzle piece :: puzzle piece,2,0,0.0,0.0,0.0
+Bend and manipulate plastic strip :: plastic strip | stationary box | table | water bottle,1,0,0.0,0.0,0.0
+Browse smartphone screen :: paper strips | smartphone | star ornaments,3,0,0.0,0.0,0.0
+Closing the door :: door | dustpan,2,0,0.0,0.0,0.0
+Counting and organizing beads :: cardboard squares | paper | pen | star-shaped beads,2,0,0.0,0.0,0.0
+Counting star beads :: paper | pen | star beads,2,0,0.0,0.0,0.0
+Cut along the marked line :: cardboard | hand | utility knife,7,0,0.0,0.0,0.0
+Cut cardboard piece :: cardboard strip | scissors,3,0,0.0,0.0,0.0
+Entering the VR training room :: doorway | person | table,3,0,0.0,0.0,0.0
+Expand paper lantern :: adhesive tape roll | cardboard box | red paper lantern,3,0,0.0,0.0,0.0
+Extract wire hangers from box :: cardboard box | wire hangers,1,0,0.0,0.0,0.0
+Fold paper lantern :: red paper honeycomb lantern,2,0,0.0,0.0,0.0
+Gesturing :: cardboard pieces | marker | pencil case | ruler | scissors,2,0,0.0,0.0,0.0
+Grasp cleaning bottle :: bowls | chopping board | cleaning fluid bottle,1,0,0.0,0.0,0.0
+Grasp lantern :: red paper honeycomb lantern,1,0,0.0,0.0,0.0
+Grasping cleaning cloth :: cleaning cloth | countertop,4,361,0.0110803324099723,1.0,0.021917808219178086
+Greeting/acknowledging participants :: person | table | vr headset,2,0,0.0,0.0,0.0
+Handle paper lantern component :: cardboard box | paper lantern component | plastic bag | red paper lantern,1,0,0.0,0.0,0.0
+Hold and bend plastic strip :: bottle | purple plastic strip | stationery box,2,0,0.0,0.0,0.0
+Hold and manipulate paper strip :: beads | mobile phone | power bank | yellow paper strip,3,0,0.0,0.0,0.0
+Hold and manipulate paper strip :: beads | paper strip | power bank | smartphone,1,0,0.0,0.0,0.0
+Hold beads :: paper | pen | power bank | smartphone | star-shaped beads,2,0,0.0,0.0,0.0
+Hold container lid :: dustpan | red plastic container lid,1,0,0.0,0.0,0.0
+Hold earbud case :: box of products | earbud case | shelf,1,0,0.0,0.0,0.0
+Hold smartphone :: quilling paper strips | small blue beads | smartphone,1,0,0.0,0.0,0.0
+Identify next cardboard piece :: cardboard pieces | marker,3,0,0.0,0.0,0.0
+Inspect shelf condition :: colleague | shelf,1,0,0.0,0.0,0.0
+Interact with smartphone :: beads | smartphone | yellow paper strip,2,0,0.0,0.0,0.0
+Interact with smartphone :: cans | cereal box | shelf | smartphone,1,0,0.0,0.0,0.0
+Interact with smartphone :: cans | shelf | smartphone,4,0,0.0,0.0,0.0
+Lift pot lid :: pot | pot lid,1,0,0.0,0.0,0.0
+Manipulate adhesive strip :: adhesive strip | paper sheets | puzzle box | smartphone | water bottle,5,0,0.0,0.0,0.0
+Manipulate bead :: bead piles | container | paper strips | yellow bead,2,0,0.0,0.0,0.0
+Manipulate bead :: bead | bead piles | pen holder | table,1,0,0.0,0.0,0.0
+Manipulate bead :: bead | bead piles | table,1,0,0.0,0.0,0.0
+Manipulate beads :: blue beads | power bank | smartphone | yellow beads,2,0,0.0,0.0,0.0
+Manipulate craft paper strips :: paper strips | scissors | smartphone,8,0,0.0,0.0,0.0
+Manipulate craft piece :: craft pieces | scissors | smartphone,2,0,0.0,0.0,0.0
+Manipulate material :: blue crafting material | hand,2,0,0.0,0.0,0.0
+Manipulate paper edge :: hands | paper cone | puzzle box | smartphone | water bottle,8,0,0.0,0.0,0.0
+Manipulate paper strip :: container with tools | table | yellow paper strip,2,0,0.0,0.0,0.0
+Manipulate paper strip :: craft materials | purple paper strip | scissors | smartphone,7,0,0.0,0.0,0.0
+Manipulate paper strip :: paper cone | paper strip | puzzle box | smartphone | water bottle,2,0,0.0,0.0,0.0
+Manipulate paper strip :: purple paper strip | quilling paper pile | smartphone,3,0,0.0,0.0,0.0
+Manipulate plastic strip :: purple plastic strip | storage bin | water bottle,6,0,0.0,0.0,0.0
+Manipulate plastic strips :: desk | plastic strips | water bottle,7,0,0.0,0.0,0.0
+Manipulate puzzle pieces :: puzzle box | puzzle mat | puzzle pieces,6,0,0.0,0.0,0.0
+Manipulate yellow strip :: beads | cell phone | pen | yellow strip,2,0,0.0,0.0,0.0
+Manipulating paper strips :: beads | blue paper strip | yellow paper strip,1,0,0.0,0.0,0.0
+Marking cardboard piece :: cardboard pieces | marker | ruler | scissors,3,0,0.0,0.0,0.0
+Move dustpan to side :: dustpan,1,2,0.0,0.0,0.0
+Move marker and adjust hand :: cardboard pieces | marker,3,0,0.0,0.0,0.0
+Move phone :: cardboard | smartphone | utility knife,13,0,0.0,0.0,0.0
+Move pot :: faucet | pot | sink,1,0,0.0,0.0,0.0
+Move smartphone :: craft beads | paper strips | smartphone,3,2,0.5,0.3333333333333333,0.4
+Observe and pause :: cardboard pieces | marker | pencil case | ruler | scissors,2,0,0.0,0.0,0.0
+Open earbud case :: earbud case,3,0,0.0,0.0,0.0
+Open folded paper lantern :: cardboard box | red pleated paper lantern,4,0,0.0,0.0,0.0
+Open paper lantern component :: paper lantern component,1,0,0.0,0.0,0.0
+Open stove pot lid :: blue bowl | cloth | faucet | red bowl | sink | soap dispenser | white bowl,2,0,0.0,0.0,0.0
+Operate smartphone :: craft paper strips | folded paper fan | paper scraps | scissors | smartphone | table,1,0,0.0,0.0,0.0
+Operate smartphone :: craft paper strips | folded paper fan | smartphone,7,1,0.0,0.0,0.0
+Operate smartphone :: paper scraps | scissors | smartphone | table,2,0,0.0,0.0,0.0
+Organize cardboard pieces :: cardboard pieces | marker | pencil case | ruler | scissors,5,0,0.0,0.0,0.0
+Pick up button :: buttons | table,3,0,0.0,0.0,0.0
+Pick up items from the shopping bag :: cardboard boxes | red shopping bag | retail shelf,3,3,0.0,0.0,0.0
+Pick up new cardboard piece :: cardboard piece | marker,2,0,0.0,0.0,0.0
+Pick up packaged paper lantern component :: cardboard box | packaged paper lantern component,1,0,0.0,0.0,0.0
+Pick up smartphone :: cans | shelf | smartphone,2,0,0.0,0.0,0.0
+Pick up star bead :: cardboard squares | mobile phone | paper | pen | power bank | star beads,1,0,0.0,0.0,0.0
+Pick up utility knife :: cardboard | utility knife,4,0,0.0,0.0,0.0
+Picking up bottle :: bottle,1,0,0.0,0.0,0.0
+Picking up crafting material :: beads | smartphone | table | yellow strip,3,0,0.0,0.0,0.0
+Place and count bead :: paper | pen | star beads,4,0,0.0,0.0,0.0
+Place another canned food on shelf :: box | canned food,2,0,0.0,0.0,0.0
+Place button :: buttons | hand | smartphone | table,3,0,0.0,0.0,0.0
+Place can on shelf :: canned food | retail shelf,3,0,0.0,0.0,0.0
+Place canned food on shelf :: canned food | store shelf,1,0,0.0,0.0,0.0
+Place cloth on floor :: cloth,1,0,0.0,0.0,0.0
+Place item on shelf :: stationery package,2,0,0.0,0.0,0.0
+Place items on the shelf :: packaged items | retail shelf,4,0,0.0,0.0,0.0
+Place material :: bead design | blue crafting material | hand,1,0,0.0,0.0,0.0
+Place phone down :: beads | pencil holder | smartphone | yellow paper strip,2,0,0.0,0.0,0.0
+Place piece into puzzle :: jigsaw puzzle | puzzle piece,4,0,0.0,0.0,0.0
+Place puzzle piece :: puzzle board | puzzle piece,3,0,0.0,0.0,0.0
+Place smartphone down :: power bank | quilling paper strips | small paper stars | smartphone,1,0,0.0,0.0,0.0
+Place smartphone down :: quilling paper strips | small paper stars | smartphone,2,0,0.0,0.0,0.0
+Place smartphone on stand :: saucepan | smartphone | smartphone stand | yellow jacket,1,0,0.0,0.0,0.0
+Place towel :: pot | towel,1,0,0.0,0.0,0.0
+Placing paper strip :: adhesive strip | paper cone | phone,4,2,0.0,0.0,0.0
+Put down smartphone :: paper quilling strips | power bank | smartphone | star beads,1,0,0.0,0.0,0.0
+Put down smartphone :: paper quilling strips | smartphone | star beads,1,2,0.0,0.0,0.0
+Put down smartphone :: power bank | smartphone,1,0,0.0,0.0,0.0
+Reach for another item :: item | packaged item | shelf,1,0,0.0,0.0,0.0
+Reach for cleaning supplies :: bowls | chopping board | cleaning fluid bottle,3,0,0.0,0.0,0.0
+Reach for next canned food :: box | canned food,1,0,0.0,0.0,0.0
+Reach for next canned food :: box | canned food | retail shelf,1,0,0.0,0.0,0.0
+Reach for next item :: packaged item | shelf,1,0,0.0,0.0,0.0
+Reach for puzzle piece :: puzzle board | puzzle box | puzzle pieces,2,0,0.0,0.0,0.0
+Reach for wire hangers :: cardboard box | wire hangers,2,0,0.0,0.0,0.0
+Reach into box :: box of cans,1,0,0.0,0.0,0.0
+Release cardboard piece and gesture :: cardboard piece | cardboard piles | marker | pouch | ruler | scissors,2,0,0.0,0.0,0.0
+Release hook :: display hook,1,0,0.0,0.0,0.0
+Release lantern :: red paper lantern,2,0,0.0,0.0,0.0
+Remove paper lantern part from packaging :: paper lantern component | plastic packaging,1,0,0.0,0.0,0.0
+Remove paper lantern part from packaging :: paper lantern | red hand fan,1,0,0.0,0.0,0.0
+Remove plastic packaging :: packaging | paper lantern component,1,0,0.0,0.0,0.0
+Reposition hand :: cardboard | utility knife,3,0,0.0,0.0,0.0
+Resume observation :: cardboard pieces | marker | pencil case | ruler | scissors,2,0,0.0,0.0,0.0
+Retrieve canned food from box :: box | canned food,1,0,0.0,0.0,0.0
+Retrieve next canned food item :: canned food | cardboard box,2,0,0.0,0.0,0.0
+Retrieving more beads :: paper | pen | star beads,2,0,0.0,0.0,0.0
+Rinse cloth in sink :: cloth | sink | water faucet,2,0,0.0,0.0,0.0
+Scroll smartphone screen :: paper stars | paper strips | smartphone,2,0,0.0,0.0,0.0
+Search for puzzle piece :: jigsaw puzzle | puzzle pieces,2,0,0.0,0.0,0.0
+Secure paper edges with adhesive :: adhesive strip | paper cone pieces | puzzle box | smartphone | water bottle,3,0,0.0,0.0,0.0
+Secure paper edges with adhesive :: adhesive strip | paper cone | paper cone pieces | puzzle box | smartphone | water bottle,1,0,0.0,0.0,0.0
+Securing paper structure :: paper cone | paper strip | puzzle box | smartphone | water bottle,4,0,0.0,0.0,0.0
+Sort and adjust button line :: buttons | smartphone | table,3,0,0.0,0.0,0.0
+Sort and arrange buttons :: buttons | smartphones | soda can | table,1,0,0.0,0.0,0.0
+Sort and arrange buttons :: buttons | smartphones | table,2,0,0.0,0.0,0.0
+Sort and count beads :: cell phone | paper | pen | star beads,1,0,0.0,0.0,0.0
+Sort and count beads :: paper | pen | star beads,2,0,0.0,0.0,0.0
+Sort beads and write count :: paper | pen | power bank | smartphone | star-shaped beads,2,0,0.0,0.0,0.0
+Sort button :: buttons | cell phone,3,0,0.0,0.0,0.0
+Sort craft items :: star-shaped craft items | table,4,2,0.0,0.0,0.0
+Sort puzzle pieces :: jigsaw puzzle pieces | table,2,0,0.0,0.0,0.0
+Sort star-shaped beads :: marker | mobile phone | paper | power bank | star-shaped beads,1,0,0.0,0.0,0.0
+Stir contents :: cooking utensil | pot,1,0,0.0,0.0,0.0
+Use smartphone :: buttons | chair | smartphone | table,1,0,0.0,0.0,0.0
+Use smartphone :: buttons | smartphone | table,2,0,0.0,0.0,0.0
+Use smartphone :: charging cable | paper strips | power bank | smartphone | star-shaped paper crafts,5,0,0.0,0.0,0.0
+Walk towards shelves :: cardboard boxes | red bin | shelving unit,4,0,0.0,0.0,0.0
+Walking towards door :: door | dustpan,1,0,0.0,0.0,0.0
+Write count on paper :: marker | mobile phone | paper | power bank | star-shaped beads,1,0,0.0,0.0,0.0

results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/predictions.csv ADDED Viewed

The diff for this file is too large to render. See raw diff

results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json CHANGED Viewed

@@ -8,7 +8,7 @@
     "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/predictions.csv"
   },
   "excluded_rows_without_true_relation": 2,
-  "generated_at_utc": "2026-06-18T20:58:29+00:00",
   "labels": [
     "Adjust canned food on shelf :: canned food | cardboard box | store shelf",
     "Adjust item on shelf :: shelf | stationery package",

     "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/predictions.csv"
   },
   "excluded_rows_without_true_relation": 2,
+  "generated_at_utc": "2026-06-18T22:52:18+00:00",
   "labels": [
     "Adjust canned food on shelf :: canned food | cardboard box | store shelf",
     "Adjust item on shelf :: shelf | stationery package",

results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/metrics.json CHANGED Viewed

@@ -8,7 +8,7 @@
     "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/predictions.csv"
   },
   "excluded_rows_without_true_relation": 18,
-  "generated_at_utc": "2026-06-18T20:58:29+00:00",
   "labels": [
     "Adjust canned food on shelf :: canned food | cardboard box | store shelf",
     "Adjust item on shelf :: hand | packaged item | shelf",

     "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/qwen3_omni_v6_lora/predictions.csv"
   },
   "excluded_rows_without_true_relation": 18,
+  "generated_at_utc": "2026-06-18T22:52:17+00:00",
   "labels": [
     "Adjust canned food on shelf :: canned food | cardboard box | store shelf",
     "Adjust item on shelf :: hand | packaged item | shelf",

results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json ADDED Viewed

	@@ -0,0 +1,28 @@

+{
+  "artifact_files": {
+    "metrics_json": "results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json",
+    "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv"
+  },
+  "caption_grounding_center_hit_rate": 0.3236607142857143,
+  "caption_grounding_iou": 0.30639899644580487,
+  "caption_grounding_within_20_frames": 0.35267857142857145,
+  "generated_at_utc": "2026-06-18T22:52:18+00:00",
+  "known_limitation": "This is an evidence-window localization score, not a candidate-set retrieval MRR.",
+  "metric_key": "caption_grounding_iou",
+  "missing_pred_evidence_window_count": 219,
+  "model_id": "cosmos3_super_reasoner",
+  "model_label": "Cosmos3-Super Reasoner",
+  "normalization_policy": "Evidence windows are frame intervals. The primary score is mean interval IoU; center-hit and within-20-frame rates are reported as diagnostics.",
+  "primary_metric": "caption_grounding_iou",
+  "primary_score": 0.30639899644580487,
+  "scope": "held_out_test_existing_verified_prediction_json",
+  "score_policy": "Derived from existing verified held-out structured JSON predictions. The task target is the evidence_window field already present in the true JSON; missing or invalid predicted evidence windows receive zero IoU.",
+  "scored_rows": 448,
+  "source_prediction_jsonl": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/eval/predictions.jsonl",
+  "status": "pass",
+  "task_id": "caption_grounding",
+  "task_label": "Language Grounding",
+  "task_number": 8,
+  "title": "Cosmos3-Super Reasoner Evidence-Window Grounding Probe",
+  "total_prediction_rows": 448
+}

results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/predictions.csv ADDED Viewed

	@@ -0,0 +1,449 @@

+id,split,episode_id,true_start_frame,true_end_frame,pred_start_frame,pred_end_frame,pred_center_frame,center_error_frames,iou
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:0,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,0.0,19.0,0.0,19.0,9.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:20,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,400.0,419.0,400.0,419.0,409.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:46,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,920.0,939.0,914.0,939.0,926.5,3.0,0.7692307692307693
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:50,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,1000.0,1019.0,0.0,19.0,9.5,1000.0,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:62,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,1240.0,1259.0,1240.0,1259.0,1249.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:70,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,1400.0,1419.0,1410.0,1419.0,1414.5,5.0,0.5
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:71,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,1420.0,1439.0,1420.0,1439.0,1429.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:90,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,1800.0,1819.0,,,,,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:100,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,2000.0,2019.0,,,,,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:120,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,2400.0,2419.0,,,,,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:126,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,2520.0,2539.0,,,,,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:127,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,2540.0,2559.0,0.0,1.0,0.5,2549.0,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:134,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,2680.0,2699.0,2680.0,2699.0,2689.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:140,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,2800.0,2819.0,,,,,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:150,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,3000.0,3019.0,0.0,1.0,0.5,3009.0,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:156,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,3120.0,3139.0,3120.0,3139.0,3129.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:166,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,3320.0,3339.0,,,,,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:170,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,3400.0,3419.0,3264.0,3419.0,3341.5,68.0,0.1282051282051282
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:184,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,3680.0,3699.0,3680.0,3699.0,3689.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:185,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,3700.0,3719.0,3681.0,3719.0,3700.0,9.5,0.5128205128205128
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:192,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,3840.0,3859.0,3840.0,3859.0,3849.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:200,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4000.0,4019.0,4000.0,4019.0,4009.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:204,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4080.0,4099.0,4080.0,4099.0,4089.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:205,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4100.0,4119.0,,,,,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:212,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4240.0,4259.0,4240.0,4259.0,4249.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:220,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4400.0,4419.0,,,,,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:224,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4480.0,4499.0,0.0,1.0,0.5,4489.0,0.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:232,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4640.0,4659.0,4640.0,4659.0,4649.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:236,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4720.0,4739.0,4720.0,4739.0,4729.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:237,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4740.0,4759.0,4740.0,4759.0,4749.5,0.0,1.0
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:240,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4800.0,4819.0,4790.0,4819.0,4804.5,5.0,0.6666666666666666
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:244,test,8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,4880.0,4899.0,4880.0,4899.0,4889.5,0.0,1.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:0,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,0.0,19.0,0.0,19.0,9.5,0.0,1.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:1,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,20.0,39.0,0.0,1.0,0.5,29.0,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:2,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,40.0,59.0,0.0,59.0,29.5,20.0,0.3333333333333333
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:3,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,60.0,79.0,0.0,1.0,0.5,69.0,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:4,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,80.0,99.0,0.0,99.0,49.5,40.0,0.2
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:5,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,100.0,119.0,0.0,119.0,59.5,50.0,0.16666666666666666
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:6,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,120.0,139.0,0.0,139.0,69.5,60.0,0.14285714285714285
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:12,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,240.0,259.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:13,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,260.0,279.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:14,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,280.0,299.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:15,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,300.0,319.0,290.0,319.0,304.5,5.0,0.6666666666666666
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:16,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,320.0,339.0,310.0,339.0,324.5,5.0,0.6666666666666666
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:17,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,340.0,359.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:18,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,360.0,379.0,360.0,379.0,369.5,0.0,1.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:19,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,380.0,399.0,380.0,399.0,389.5,0.0,1.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:20,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,400.0,419.0,393.0,419.0,406.0,3.5,0.7407407407407407
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:21,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,420.0,439.0,419.0,439.0,429.0,0.5,0.9523809523809523
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:22,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,440.0,459.0,430.0,459.0,444.5,5.0,0.6666666666666666
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:23,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,460.0,479.0,460.0,479.0,469.5,0.0,1.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:24,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,480.0,499.0,480.0,499.0,489.5,0.0,1.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:25,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,500.0,519.0,493.0,519.0,506.0,3.5,0.7407407407407407
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:26,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,520.0,539.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:250,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5000.0,5019.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:251,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5020.0,5039.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:252,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5040.0,5059.0,5040.0,5059.0,5049.5,0.0,1.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:253,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5060.0,5079.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:254,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5080.0,5099.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:255,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5100.0,5119.0,5090.0,5119.0,5104.5,5.0,0.6666666666666666
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:256,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5120.0,5139.0,5114.0,5139.0,5126.5,3.0,0.7692307692307693
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:294,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5880.0,5899.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:295,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5900.0,5919.0,,,,,0.0
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:296,test,a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,5920.0,5939.0,0.0,1.0,0.5,5929.0,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:0,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,0.0,19.0,0.0,1.0,0.5,9.0,0.1
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:1,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,20.0,39.0,,,,,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:16,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,320.0,339.0,0.0,1.0,0.5,329.0,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:17,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,340.0,359.0,0.0,1.0,0.5,349.0,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:20,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,400.0,419.0,380.0,419.0,399.5,10.0,0.5
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:21,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,420.0,439.0,,,,,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:46,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,920.0,939.0,,,,,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:47,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,940.0,959.0,,,,,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:50,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,1000.0,1019.0,0.0,1.0,0.5,1009.0,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:58,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,1160.0,1179.0,1160.0,1179.0,1169.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:59,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,1180.0,1199.0,0.0,0.0,0.0,1189.5,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:64,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,1280.0,1299.0,1280.0,1299.0,1289.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:65,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,1300.0,1319.0,1300.0,1319.0,1309.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:70,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,1400.0,1419.0,1398.0,1419.0,1408.5,1.0,0.9090909090909091
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:76,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,1520.0,1539.0,1520.0,1539.0,1529.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:77,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,1540.0,1559.0,,,,,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:100,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,2000.0,2019.0,1965.0,2019.0,1992.0,17.5,0.36363636363636365
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:101,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,2020.0,2039.0,0.0,1.0,0.5,2029.0,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:126,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,2520.0,2539.0,2518.0,2539.0,2528.5,1.0,0.9090909090909091
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:138,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,2760.0,2779.0,0.0,1.0,0.5,2769.0,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:140,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,2800.0,2819.0,0.0,19.0,9.5,2800.0,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:141,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,2820.0,2839.0,2820.0,2839.0,2829.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:150,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,3000.0,3019.0,3000.0,3019.0,3009.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:151,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,3020.0,3039.0,3020.0,3039.0,3029.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:164,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,3280.0,3299.0,3280.0,3299.0,3289.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:165,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,3300.0,3319.0,0.0,19.0,9.5,3300.0,0.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:172,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,3440.0,3459.0,3380.0,3459.0,3419.5,30.0,0.25
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:173,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,3460.0,3479.0,3460.0,3479.0,3469.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:182,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,3640.0,3659.0,3639.0,3659.0,3649.0,0.5,0.9523809523809523
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:183,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,3660.0,3679.0,3660.0,3679.0,3669.5,0.0,1.0
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:226,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,4520.0,4539.0,4518.0,4539.0,4528.5,1.0,0.9090909090909091
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1:qa:244,test,33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,4880.0,4899.0,4880.0,4899.0,4889.5,0.0,1.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:0,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,0.0,19.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:1,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,20.0,39.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:2,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,40.0,59.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:26,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,520.0,539.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:27,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,540.0,559.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:30,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,600.0,619.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:31,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,620.0,639.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:70,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,1400.0,1419.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:71,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,1420.0,1439.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:72,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,1440.0,1459.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:90,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,1800.0,1819.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:91,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,1820.0,1839.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:110,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,2200.0,2219.0,2194.0,2219.0,2206.5,3.0,0.7692307692307693
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:111,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,2220.0,2239.0,2220.0,2239.0,2229.5,0.0,1.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:112,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,2240.0,2259.0,2240.0,2259.0,2249.5,0.0,1.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:170,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,3400.0,3419.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:171,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,3420.0,3439.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:172,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,3440.0,3459.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:230,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,4600.0,4619.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:231,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,4620.0,4639.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:276,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5520.0,5539.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:277,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5540.0,5559.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:278,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5560.0,5579.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:282,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5640.0,5659.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:283,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5660.0,5679.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:284,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5680.0,5699.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:290,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5800.0,5819.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:291,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5820.0,5839.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:292,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5840.0,5859.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:293,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5860.0,5879.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:294,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5880.0,5899.0,,,,,0.0
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2:qa:295,test,9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,5900.0,5919.0,,,,,0.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:0,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,0.0,19.0,0.0,19.0,9.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:12,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,240.0,259.0,240.0,259.0,249.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:64,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,1280.0,1299.0,1280.0,1299.0,1289.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:65,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,1300.0,1319.0,1300.0,1319.0,1309.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:70,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,1400.0,1419.0,1395.0,1419.0,1407.0,2.5,0.8
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:71,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,1420.0,1439.0,1420.0,1439.0,1429.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:82,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,1640.0,1659.0,1640.0,1659.0,1649.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:90,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,1800.0,1819.0,1796.0,1819.0,1807.5,2.0,0.8333333333333334
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:91,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,1820.0,1839.0,1820.0,1839.0,1829.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:106,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,2120.0,2139.0,2120.0,2139.0,2129.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:140,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,2800.0,2819.0,2698.0,2819.0,2758.5,51.0,0.16393442622950818
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:141,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,2820.0,2839.0,2816.0,2839.0,2827.5,2.0,0.8333333333333334
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:154,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,3080.0,3099.0,3080.0,3099.0,3089.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:160,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,3200.0,3219.0,0.0,19.0,9.5,3200.0,0.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:166,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,3320.0,3339.0,3320.0,3339.0,3329.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:167,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,3340.0,3359.0,3340.0,3359.0,3349.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:190,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,3800.0,3819.0,3795.0,3819.0,3807.0,2.5,0.8
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:191,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,3820.0,3839.0,3819.0,3839.0,3829.0,0.5,0.9523809523809523
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:198,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,3960.0,3979.0,3960.0,3979.0,3969.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:199,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,3980.0,3999.0,3980.0,3999.0,3989.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:206,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4120.0,4139.0,4116.0,4139.0,4127.5,2.0,0.8333333333333334
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:207,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4140.0,4159.0,4140.0,4159.0,4149.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:218,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4360.0,4379.0,,,,,0.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:219,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4380.0,4399.0,4380.0,4399.0,4389.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:220,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4400.0,4419.0,,,,,0.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:221,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4420.0,4439.0,4420.0,4439.0,4429.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:236,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4720.0,4739.0,,,,,0.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:237,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4740.0,4759.0,4740.0,4759.0,4749.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:240,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4800.0,4819.0,4799.0,4819.0,4809.0,0.5,0.9523809523809523
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:241,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4820.0,4839.0,0.0,1.0,0.5,4829.0,0.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:244,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4880.0,4899.0,4880.0,4899.0,4889.5,0.0,1.0
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5:qa:245,test,34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,4900.0,4919.0,0.0,19.0,9.5,4900.0,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:0,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,0.0,19.0,0.0,1.0,0.5,9.0,0.1
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:1,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,20.0,39.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:2,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,40.0,59.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:40,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,800.0,819.0,782.0,819.0,800.5,9.0,0.5263157894736842
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:41,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,820.0,839.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:42,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,840.0,859.0,840.0,859.0,849.5,0.0,1.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:43,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,860.0,879.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:70,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1400.0,1419.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:71,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1420.0,1439.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:72,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1440.0,1459.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:73,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1460.0,1479.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:90,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1800.0,1819.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:91,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1820.0,1839.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:92,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1840.0,1859.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:94,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1880.0,1899.0,0.0,1.0,0.5,1889.0,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:95,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1900.0,1919.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:96,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1920.0,1939.0,1920.0,1939.0,1929.5,0.0,1.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:97,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,1940.0,1959.0,0.0,1.0,0.5,1949.0,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:114,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2280.0,2299.0,2280.0,2299.0,2289.5,0.0,1.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:115,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2300.0,2319.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:116,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2320.0,2339.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:120,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2400.0,2419.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:121,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2420.0,2439.0,2390.0,2439.0,2414.5,15.0,0.4
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:122,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2440.0,2459.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:123,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2460.0,2479.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:144,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2880.0,2899.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:145,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2900.0,2919.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:146,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2920.0,2939.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:147,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,2940.0,2959.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:190,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,3800.0,3819.0,,,,,0.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:191,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,3820.0,3839.0,3820.0,3839.0,3829.5,0.0,1.0
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2:qa:192,test,b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,3840.0,3859.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:0,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,0.0,19.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:1,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,20.0,39.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:20,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,400.0,419.0,400.0,419.0,409.5,0.0,1.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:21,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,420.0,439.0,0.0,1.0,0.5,429.0,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:30,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,600.0,619.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:31,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,620.0,639.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:40,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,800.0,819.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:41,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,820.0,839.0,0.0,1.0,0.5,829.0,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:70,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,1400.0,1419.0,1380.0,1419.0,1399.5,10.0,0.5
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:71,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,1420.0,1439.0,1420.0,1439.0,1429.5,0.0,1.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:170,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,3400.0,3419.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:171,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,3420.0,3439.0,3420.0,3439.0,3429.5,0.0,1.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:176,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,3520.0,3539.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:182,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,3640.0,3659.0,0.0,1.0,0.5,3649.0,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:190,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,3800.0,3819.0,0.0,1.0,0.5,3809.0,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:191,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,3820.0,3839.0,0.0,1.0,0.5,3829.0,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:200,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4000.0,4019.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:201,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4020.0,4039.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:214,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4280.0,4299.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:215,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4300.0,4319.0,4285.0,4319.0,4302.0,7.5,0.5714285714285714
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:220,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4400.0,4419.0,4399.0,4419.0,4409.0,0.5,0.9523809523809523
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:221,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4420.0,4439.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:230,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4600.0,4619.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:231,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4620.0,4639.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:240,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,4800.0,4819.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:250,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,5000.0,5019.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:258,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,5160.0,5179.0,0.0,1.0,0.5,5169.0,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:259,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,5180.0,5199.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:270,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,5400.0,5419.0,5400.0,5419.0,5409.5,0.0,1.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:271,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,5420.0,5439.0,,,,,0.0
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:290,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,5800.0,5819.0,5785.0,5819.0,5802.0,7.5,0.5714285714285714
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2:qa:291,test,ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,5820.0,5839.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:0,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,0.0,19.0,0.0,2.0,1.0,8.5,0.15
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:1,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,20.0,39.0,0.0,1.0,0.5,29.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:2,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,40.0,59.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:14,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,280.0,299.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:15,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,300.0,319.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:16,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,320.0,339.0,320.0,339.0,329.5,0.0,1.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:20,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,400.0,419.0,380.0,419.0,399.5,10.0,0.5
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:21,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,420.0,439.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:22,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,440.0,459.0,440.0,459.0,449.5,0.0,1.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:32,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,640.0,659.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:33,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,660.0,679.0,660.0,679.0,669.5,0.0,1.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:34,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,680.0,699.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:50,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,1000.0,1019.0,0.0,1.6,0.8,1008.7,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:51,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,1020.0,1039.0,1020.0,1039.0,1029.5,0.0,1.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:70,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,1400.0,1419.0,0.0,1.0,0.5,1409.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:71,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,1420.0,1439.0,0.0,1.0,0.5,1429.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:72,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,1440.0,1459.0,0.0,15.0,7.5,1442.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:100,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,2000.0,2019.0,0.0,19.0,9.5,2000.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:101,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,2020.0,2039.0,0.0,1.0,0.5,2029.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:102,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,2040.0,2059.0,0.0,1.0,0.5,2049.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:132,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,2640.0,2659.0,0.0,19.0,9.5,2640.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:133,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,2660.0,2679.0,2660.0,2679.0,2669.5,0.0,1.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:134,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,2680.0,2699.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:184,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,3680.0,3699.0,3680.0,3699.0,3689.5,0.0,1.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:185,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,3700.0,3719.0,0.0,19.0,9.5,3700.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:186,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,3720.0,3739.0,3720.0,3739.0,3729.5,0.0,1.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:200,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,4000.0,4019.0,0.0,1.0,0.5,4009.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:201,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,4020.0,4039.0,0.0,1.0,0.5,4029.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:202,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,4040.0,4059.0,0.0,1.0,0.5,4049.0,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:220,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,4400.0,4419.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:221,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,4420.0,4439.0,,,,,0.0
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1:qa:222,test,4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,4440.0,4459.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:0,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,0.0,19.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:1,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,20.0,39.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:2,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,40.0,59.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:8,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,160.0,179.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:9,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,180.0,199.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:10,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,200.0,219.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:20,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,400.0,419.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:21,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,420.0,439.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:22,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,440.0,459.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:50,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,1000.0,1019.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:51,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,1020.0,1039.0,1020.0,1039.0,1029.5,0.0,1.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:52,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,1040.0,1059.0,1040.0,1059.0,1049.5,0.0,1.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:90,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,1800.0,1819.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:91,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,1820.0,1839.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:92,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,1840.0,1859.0,0.0,0.0,0.0,1849.5,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:100,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2000.0,2019.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:101,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2020.0,2039.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:102,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2040.0,2059.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:120,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2400.0,2419.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:121,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2420.0,2439.0,2420.0,2439.0,2429.5,0.0,1.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:140,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2800.0,2819.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:141,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2820.0,2839.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:144,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2880.0,2899.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:145,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2900.0,2919.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:146,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,2920.0,2939.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:150,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,3000.0,3019.0,3000.0,3019.0,3009.5,0.0,1.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:151,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,3020.0,3039.0,3020.0,3039.0,3029.5,0.0,1.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:160,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,3200.0,3219.0,3200.0,3219.0,3209.5,0.0,1.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:161,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,3220.0,3239.0,3220.0,3239.0,3229.5,0.0,1.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:162,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,3240.0,3259.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:190,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,3800.0,3819.0,,,,,0.0
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6:qa:191,test,5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,3820.0,3839.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:0,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,0.0,19.0,0.0,0.0,0.0,9.5,0.05
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:20,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,400.0,419.0,380.0,419.0,399.5,10.0,0.5
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:21,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,420.0,439.0,420.0,439.0,429.5,0.0,1.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:28,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,560.0,579.0,560.0,579.0,569.5,0.0,1.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:29,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,580.0,599.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:34,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,680.0,699.0,680.0,699.0,689.5,0.0,1.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:35,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,700.0,719.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:40,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,800.0,819.0,794.0,819.0,806.5,3.0,0.7692307692307693
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:41,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,820.0,839.0,819.0,839.0,829.0,0.5,0.9523809523809523
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:70,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,1400.0,1419.0,1400.0,1419.0,1409.5,0.0,1.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:71,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,1420.0,1439.0,1420.0,1439.0,1429.5,0.0,1.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:74,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,1480.0,1499.0,1480.0,1499.0,1489.5,0.0,1.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:75,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,1500.0,1519.0,1494.0,1519.0,1506.5,3.0,0.7692307692307693
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:82,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,1640.0,1659.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:83,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,1660.0,1679.0,1660.0,1679.0,1669.5,0.0,1.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:90,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,1800.0,1819.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:91,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,1820.0,1839.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:120,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,2400.0,2419.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:121,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,2420.0,2439.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:150,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,3000.0,3019.0,0.0,19.0,9.5,3000.0,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:151,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,3020.0,3039.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:190,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,3800.0,3819.0,0.0,19.0,9.5,3800.0,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:191,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,3820.0,3839.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:220,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,4400.0,4419.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:221,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,4420.0,4439.0,4386.0,4439.0,4412.5,17.0,0.37037037037037035
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:240,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,4800.0,4819.0,0.0,19.0,9.5,4800.0,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:246,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,4920.0,4939.0,0.0,1.0,0.5,4929.0,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:247,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,4940.0,4959.0,4940.0,4959.0,4949.5,0.0,1.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:250,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,5000.0,5019.0,,,,,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:251,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,5020.0,5039.0,0.0,1.0,0.5,5029.0,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:296,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,5920.0,5939.0,0.0,1.0,0.5,5929.0,0.0
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1:qa:297,test,b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,5940.0,5959.0,0.0,19.0,9.5,5940.0,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:0,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,0.0,19.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:1,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,20.0,39.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:2,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,40.0,59.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:20,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,400.0,419.0,380.0,419.0,399.5,10.0,0.5
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:21,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,420.0,439.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:22,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,440.0,459.0,440.0,459.0,449.5,0.0,1.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:23,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,460.0,479.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:50,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,1000.0,1019.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:51,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,1020.0,1039.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:52,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,1040.0,1059.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:53,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,1060.0,1079.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:70,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,1400.0,1419.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:71,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,1420.0,1439.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:72,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,1440.0,1459.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:100,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2000.0,2019.0,1991.0,2019.0,2005.0,4.5,0.6896551724137931
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:101,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2020.0,2039.0,2019.0,2039.0,2029.0,0.5,0.9523809523809523
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:102,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2040.0,2059.0,2040.0,2059.0,2049.5,0.0,1.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:108,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2160.0,2179.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:109,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2180.0,2199.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:110,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2200.0,2219.0,2199.0,2219.0,2209.0,0.5,0.9523809523809523
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:120,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2400.0,2419.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:121,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2420.0,2439.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:122,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,2440.0,2459.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:150,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,3000.0,3019.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:151,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,3020.0,3039.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:152,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,3040.0,3059.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:170,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,3400.0,3419.0,0.0,19.0,9.5,3400.0,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:171,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,3420.0,3439.0,0.0,1.0,0.5,3429.0,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:172,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,3440.0,3459.0,0.0,1.0,0.5,3449.0,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:200,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,4000.0,4019.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:201,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,4020.0,4039.0,,,,,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4:qa:202,test,877779cd-25f3-4293-a3c4-39067dd9558c__ep4,4040.0,4059.0,4040.0,4059.0,4049.5,0.0,1.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:0,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,0.0,19.0,0.0,1.0,0.5,9.0,0.1
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:1,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,20.0,39.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:2,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,40.0,59.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:10,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,200.0,219.0,180.0,219.0,199.5,10.0,0.5
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:11,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,220.0,239.0,220.0,239.0,229.5,0.0,1.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:12,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,240.0,259.0,0.0,10.0,5.0,244.5,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:13,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,260.0,279.0,0.0,1.0,0.5,269.0,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:20,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,400.0,419.0,400.0,419.0,409.5,0.0,1.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:21,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,420.0,439.0,420.0,439.0,429.5,0.0,1.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:22,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,440.0,459.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:28,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,560.0,579.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:29,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,580.0,599.0,580.0,599.0,589.5,0.0,1.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:30,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,600.0,619.0,608.0,619.0,613.5,4.0,0.6
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:40,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,800.0,819.0,800.0,819.0,809.5,0.0,1.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:41,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,820.0,839.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:42,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,840.0,859.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:50,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1000.0,1019.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:51,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1020.0,1039.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:52,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1040.0,1059.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:70,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1400.0,1419.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:71,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1420.0,1439.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:72,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1440.0,1459.0,1440.0,1459.0,1449.5,0.0,1.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:90,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1800.0,1819.0,0.0,19.0,9.5,1800.0,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:91,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1820.0,1839.0,0.0,15.0,7.5,1822.0,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:92,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,1840.0,1859.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:100,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,2000.0,2019.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:101,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,2020.0,2039.0,0.0,10.0,5.0,2024.5,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:102,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,2040.0,2059.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:103,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,2060.0,2079.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:140,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,2800.0,2819.0,2666.0,2819.0,2742.5,67.0,0.12987012987012986
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:141,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,2820.0,2839.0,,,,,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1:qa:142,test,1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,2840.0,2859.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:0,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,0.0,19.0,0.0,1.0,0.5,9.0,0.1
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:1,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,20.0,39.0,0.0,19.0,9.5,20.0,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:2,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,40.0,59.0,0.0,59.0,29.5,20.0,0.3333333333333333
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:3,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,60.0,79.0,0.0,1.0,0.5,69.0,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:4,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,80.0,99.0,0.0,99.0,49.5,40.0,0.2
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:40,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,800.0,819.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:41,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,820.0,839.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:42,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,840.0,859.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:43,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,860.0,879.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:44,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,880.0,899.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:60,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1200.0,1219.0,1199.0,1219.0,1209.0,0.5,0.9523809523809523
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:61,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1220.0,1239.0,1220.0,1239.0,1229.5,0.0,1.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:62,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1240.0,1259.0,1240.0,1259.0,1249.5,0.0,1.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:63,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1260.0,1279.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:90,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1800.0,1819.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:91,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1820.0,1839.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:92,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1840.0,1859.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:93,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1860.0,1879.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:94,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,1880.0,1899.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:146,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,2920.0,2939.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:147,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,2940.0,2959.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:148,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,2960.0,2979.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:149,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,2980.0,2999.0,,,,,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:170,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,3400.0,3419.0,3399.0,3419.0,3409.0,0.5,0.9523809523809523
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:171,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,3420.0,3439.0,0.0,39.0,19.5,3410.0,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:172,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,3440.0,3459.0,0.0,1.0,0.5,3449.0,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:173,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,3460.0,3479.0,0.0,1.0,0.5,3469.0,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:174,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,3480.0,3499.0,3480.0,3499.0,3489.5,0.0,1.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:240,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,4800.0,4819.0,0.0,1.0,0.5,4809.0,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:241,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,4820.0,4839.0,0.0,1.0,0.5,4829.0,0.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:242,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,4840.0,4859.0,4840.0,4859.0,4849.5,0.0,1.0
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2:qa:243,test,ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,4860.0,4879.0,0.0,1.0,0.5,4869.0,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:0,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,0.0,19.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:1,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,20.0,39.0,0.0,2.0,1.0,28.5,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:14,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,280.0,299.0,280.0,299.0,289.5,0.0,1.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:15,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,300.0,319.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:50,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,1000.0,1019.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:100,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,2000.0,2019.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:101,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,2020.0,2039.0,2014.0,2039.0,2026.5,3.0,0.7692307692307693
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:120,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,2400.0,2419.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:121,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,2420.0,2439.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:128,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,2560.0,2579.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:129,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,2580.0,2599.0,2580.0,2599.0,2589.5,0.0,1.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:150,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,3000.0,3019.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:164,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,3280.0,3299.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:180,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,3600.0,3619.0,3598.0,3619.0,3608.5,1.0,0.9090909090909091
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:181,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,3620.0,3639.0,3611.0,3639.0,3625.0,4.5,0.6896551724137931
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:200,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4000.0,4019.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:201,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4020.0,4039.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:212,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4240.0,4259.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:213,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4260.0,4279.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:220,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4400.0,4419.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:221,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4420.0,4439.0,4385.0,4439.0,4412.0,17.5,0.36363636363636365
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:232,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4640.0,4659.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:233,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4660.0,4679.0,4654.0,4679.0,4666.5,3.0,0.7692307692307693
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:240,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4800.0,4819.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:244,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4880.0,4899.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:245,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,4900.0,4919.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:250,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,5000.0,5019.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:251,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,5020.0,5039.0,5020.0,5039.0,5029.5,0.0,1.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:256,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,5120.0,5139.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:257,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,5140.0,5159.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:264,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,5280.0,5299.0,,,,,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3:qa:265,test,b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,5300.0,5319.0,,,,,0.0

results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/metrics.json CHANGED Viewed

@@ -5,7 +5,7 @@
     "per_class_metrics_csv": "results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/per_class_metrics.csv",
     "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/predictions.csv"
   },
-  "generated_at_utc": "2026-06-18T20:58:30+00:00",
   "known_limitation": "This is a retrieval-derived future-subtask score. It is not a separately prompted generative subtask head, but it is target-backed by verified future-record retrieval outputs.",
   "labels": [
     "Adjust lantern string and handle components",

     "per_class_metrics_csv": "results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/per_class_metrics.csv",
     "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/next_subtask_forecast/cosmos3_nano_future_window/predictions.csv"
   },
+  "generated_at_utc": "2026-06-18T22:52:18+00:00",
   "known_limitation": "This is a retrieval-derived future-subtask score. It is not a separately prompted generative subtask head, but it is target-backed by verified future-record retrieval outputs.",
   "labels": [
     "Adjust lantern string and handle components",

results/omni_finetune/model_output_task_probes_20260616/object_set_forecast/cosmos3_nano_future_window/metrics.json CHANGED Viewed

@@ -6,7 +6,7 @@
   "exact_match": 0.013227513227513227,
   "false_negative_objects": 1127.0,
   "false_positive_objects": 747.0,
-  "generated_at_utc": "2026-06-18T20:58:30+00:00",
   "known_limitation": "This is a retrieval-derived object-set score, not a separately prompted object-list generation head.",
   "metric_key": "object_set_forecast_micro_f1",
   "micro_f1": 0.01781970649895178,

   "exact_match": 0.013227513227513227,
   "false_negative_objects": 1127.0,
   "false_positive_objects": 747.0,
+  "generated_at_utc": "2026-06-18T22:52:18+00:00",
   "known_limitation": "This is a retrieval-derived object-set score, not a separately prompted object-list generation head.",
   "metric_key": "object_set_forecast_micro_f1",
   "micro_f1": 0.01781970649895178,

results/omni_finetune/model_output_task_probes_20260616/summary.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-  "generated_at_utc": "2026-06-18T20:58:30+00:00",
   "methods": {
     "cosmos3_nano_future_window": {
       "label": "Cosmos3-Nano Future Window",
@@ -7,6 +7,12 @@
       "source_prediction_jsonl": "results/omni_finetune/verified_public/xperience10m_cosmos3_nano_128ep_future_window_h5_compat_adapter_eval_test_full/eval/future_predictions.jsonl",
       "status": "scored",
       "tasks": {
         "long_horizon_next_action": {
           "horizon_windows": 5,
           "long_horizon_next_action_accuracy": 0.007936507936507936,
@@ -41,9 +47,7 @@
           "within_20_frames": 0.6666666666666666
         }
       },
-      "unsupported_tasks": {
-        "action_object_relation": "verified future-window predictions do not contain object-set fields"
-      }
     },
     "cosmos3_super_reasoner": {
       "label": "Cosmos3-Super Reasoner",
@@ -58,6 +62,13 @@
           "source_metrics_json": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
           "valid_pred_relation_rate": 0.49327354260089684
         },
         "long_horizon_next_action": {
           "long_horizon_next_action_accuracy": 0.03794642857142857,
           "long_horizon_next_action_macro_f1": 0.008807588075880758,
@@ -91,10 +102,11 @@
     }
   },
   "scope": "Task-specific scoring from existing verified held-out model outputs. No new model inference, training, or target backfilling is performed.",
-  "scored_method_task_count_added": 9,
   "status": "pass",
   "task_ids_added_to_matrix": [
     "action_object_relation",
     "long_horizon_next_action",
     "modality_reconstruction",
     "next_subtask_forecast",

 {
+  "generated_at_utc": "2026-06-18T22:52:18+00:00",
   "methods": {
     "cosmos3_nano_future_window": {
       "label": "Cosmos3-Nano Future Window",
       "source_prediction_jsonl": "results/omni_finetune/verified_public/xperience10m_cosmos3_nano_128ep_future_window_h5_compat_adapter_eval_test_full/eval/future_predictions.jsonl",
       "status": "scored",
       "tasks": {
+        "action_object_relation": {
+          "action_object_relation_accuracy": 0.013297872340425532,
+          "action_object_relation_macro_f1": 0.002794157670325683,
+          "scored_rows": 376,
+          "source_metrics_json": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_nano_future_window/metrics.json"
+        },
         "long_horizon_next_action": {
           "horizon_windows": 5,
           "long_horizon_next_action_accuracy": 0.007936507936507936,
           "within_20_frames": 0.6666666666666666
         }
       },
+      "unsupported_tasks": {}
     },
     "cosmos3_super_reasoner": {
       "label": "Cosmos3-Super Reasoner",
           "source_metrics_json": "results/omni_finetune/model_output_task_probes_20260616/action_object_relation/cosmos3_super_reasoner/metrics.json",
           "valid_pred_relation_rate": 0.49327354260089684
         },
+        "caption_grounding": {
+          "caption_grounding_center_hit_rate": 0.3236607142857143,
+          "caption_grounding_iou": 0.30639899644580487,
+          "missing_pred_evidence_window_count": 219,
+          "scored_rows": 448,
+          "source_metrics_json": "results/omni_finetune/model_output_task_probes_20260616/caption_grounding/cosmos3_super_reasoner/metrics.json"
+        },
         "long_horizon_next_action": {
           "long_horizon_next_action_accuracy": 0.03794642857142857,
           "long_horizon_next_action_macro_f1": 0.008807588075880758,
     }
   },
   "scope": "Task-specific scoring from existing verified held-out model outputs. No new model inference, training, or target backfilling is performed.",
+  "scored_method_task_count_added": 11,
   "status": "pass",
   "task_ids_added_to_matrix": [
     "action_object_relation",
+    "caption_grounding",
     "long_horizon_next_action",
     "modality_reconstruction",
     "next_subtask_forecast",

results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/metrics.json CHANGED Viewed

@@ -4,7 +4,7 @@
     "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/predictions.csv"
   },
   "cap_frames": 200,
-  "generated_at_utc": "2026-06-18T20:58:30+00:00",
   "known_limitation": "This is a derived future-window action-sequence probe, not evidence of a separately trained scalar time-regression head. It is included because task 20's boundary target is deterministically derivable once future-window action predictions are verified.",
   "metric_direction": "lower",
   "metric_key": "time_to_transition_mae",

     "predictions_csv": "results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_nano_future_window/predictions.csv"
   },
   "cap_frames": 200,
+  "generated_at_utc": "2026-06-18T22:52:18+00:00",
   "known_limitation": "This is a derived future-window action-sequence probe, not evidence of a separately trained scalar time-regression head. It is included because task 20's boundary target is deterministically derivable once future-window action predictions are verified.",
   "metric_direction": "lower",
   "metric_key": "time_to_transition_mae",

results/omni_finetune/model_output_task_probes_20260616/time_to_transition/cosmos3_super_reasoner/metrics.json CHANGED Viewed

@@ -5,7 +5,7 @@
   },
   "cap_frames": 200,
   "excluded_rows_without_true_action": 0,
-  "generated_at_utc": "2026-06-18T20:58:30+00:00",
   "known_limitation": "This is a derived action-sequence probe, not evidence of a separately trained time-regression head. It is included because task 20's target is deterministically derivable from a sequence of action labels.",
   "metric_direction": "lower",
   "metric_key": "time_to_transition_mae",

   },
   "cap_frames": 200,
   "excluded_rows_without_true_action": 0,
+  "generated_at_utc": "2026-06-18T22:52:18+00:00",
   "known_limitation": "This is a derived action-sequence probe, not evidence of a separately trained time-regression head. It is included because task 20's target is deterministically derivable from a sequence of action labels.",
   "metric_direction": "lower",
   "metric_key": "time_to_transition_mae",

scripts/build_unified_task_model_radar.py CHANGED Viewed

@@ -71,6 +71,9 @@ QWEN_ACTION_OBJECT_METRICS_PATH = (
 COSMOS_SUPER_ACTION_OBJECT_METRICS_PATH = (
     MODEL_OUTPUT_TASK_PROBE_DIR / "action_object_relation/cosmos3_super_reasoner/metrics.json"
 )
 COSMOS_SUPER_TIME_TO_TRANSITION_METRICS_PATH = (
     MODEL_OUTPUT_TASK_PROBE_DIR / "time_to_transition/cosmos3_super_reasoner/metrics.json"
 )
@@ -89,6 +92,9 @@ COSMOS_NANO_MODALITY_RECONSTRUCTION_METRICS_PATH = (
 COSMOS_NANO_OBJECT_SET_METRICS_PATH = (
     MODEL_OUTPUT_TASK_PROBE_DIR / "object_set_forecast/cosmos3_nano_future_window/metrics.json"
 )
 COSMOS_NANO_TIME_TO_TRANSITION_METRICS_PATH = (
     MODEL_OUTPUT_TASK_PROBE_DIR / "time_to_transition/cosmos3_nano_future_window/metrics.json"
 )
@@ -231,6 +237,10 @@ FOUNDATION_TASK_METRICS = {
     "action_object_relation": {
         "qwen3_omni_v6_lora": "action_object_relation_macro_f1",
         "cosmos3_super_reasoner": "action_object_relation_macro_f1",
     },
     "long_horizon_next_action": {
         "cosmos3_super_reasoner": "long_horizon_next_action_macro_f1",
@@ -263,6 +273,7 @@ FOUNDATION_METRIC_PATHS = {
 FOUNDATION_METRIC_SOURCE_OVERRIDES = {
     ("qwen3_omni_v6_lora", "action_object_relation"): QWEN_ACTION_OBJECT_METRICS_PATH,
     ("cosmos3_super_reasoner", "action_object_relation"): COSMOS_SUPER_ACTION_OBJECT_METRICS_PATH,
     ("qwen3_omni_v6_lora", "caption_grounding"): QWEN_FUTURE_TASK_METRIC_PATHS["caption_grounding"],
     ("qwen3_omni_v6_lora", "cross_modal_retrieval"): QWEN_FUTURE_TASK_METRIC_PATHS["cross_modal_retrieval"],
     ("qwen3_omni_v6_lora", "temporal_order"): QWEN_FUTURE_TASK_METRIC_PATHS["temporal_order"],
@@ -275,6 +286,7 @@ FOUNDATION_METRIC_SOURCE_OVERRIDES = {
     ("cosmos3_nano_future_window", "long_horizon_next_action"): COSMOS_NANO_LONG_HORIZON_METRICS_PATH,
     ("cosmos3_nano_future_window", "next_subtask_forecast"): COSMOS_NANO_NEXT_SUBTASK_METRICS_PATH,
     ("cosmos3_nano_future_window", "modality_reconstruction"): COSMOS_NANO_MODALITY_RECONSTRUCTION_METRICS_PATH,
     ("cosmos3_nano_future_window", "object_set_forecast"): COSMOS_NANO_OBJECT_SET_METRICS_PATH,
     ("cosmos3_nano_future_window", "time_to_transition"): COSMOS_NANO_TIME_TO_TRANSITION_METRICS_PATH,
     ("cosmos3_super_reasoner", "long_horizon_next_action"): COSMOS_SUPER_LONG_HORIZON_METRICS_PATH,
@@ -312,8 +324,8 @@ METHOD_DETAILS = {
     "raw128_simple": "128-episode 4430-dim sensor NPZ simple heads; tasks 15/19 use compact proxies.",
     "raw128_neural_mlp": "128-episode 4430-dim sensor NPZ MLP heads; tasks 15/19 use compact proxies.",
     "qwen3_omni_v6_lora": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
-    "cosmos3_super_reasoner": "Verified Cosmos3-Super base-weight Reasoner JSON-task evaluation, plus task 16 and a derived task-20 action-boundary timing probe scored from existing verified JSON.",
-    "cosmos3_nano_future_window": "Verified Cosmos3-Nano future-window compatibility metrics, plus tasks 10/13/14/17 and a derived task-20 boundary timing probe scored from existing held-out future-window artifacts.",
 }
 PROXY_TASK_IDS = {"interaction_text_prediction", "camera_view_sync_retrieval"}
@@ -669,11 +681,13 @@ def build_payload() -> dict[str, Any]:
     cosmos_fd = read_json(COSMOS_SUPER_FD_METRICS_PATH)
     qwen.update(read_json(QWEN_ACTION_OBJECT_METRICS_PATH))
     cosmos_super.update(read_json(COSMOS_SUPER_ACTION_OBJECT_METRICS_PATH))
     cosmos_super.update(read_json(COSMOS_SUPER_LONG_HORIZON_METRICS_PATH))
     cosmos_super.update(read_json(COSMOS_SUPER_TIME_TO_TRANSITION_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_LONG_HORIZON_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_NEXT_SUBTASK_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_MODALITY_RECONSTRUCTION_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_OBJECT_SET_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_TIME_TO_TRANSITION_METRICS_PATH))
     foundation_task_metrics = foundation_task_metric_mapping(qwen)

 COSMOS_SUPER_ACTION_OBJECT_METRICS_PATH = (
     MODEL_OUTPUT_TASK_PROBE_DIR / "action_object_relation/cosmos3_super_reasoner/metrics.json"
 )
+COSMOS_SUPER_CAPTION_GROUNDING_METRICS_PATH = (
+    MODEL_OUTPUT_TASK_PROBE_DIR / "caption_grounding/cosmos3_super_reasoner/metrics.json"
+)
 COSMOS_SUPER_TIME_TO_TRANSITION_METRICS_PATH = (
     MODEL_OUTPUT_TASK_PROBE_DIR / "time_to_transition/cosmos3_super_reasoner/metrics.json"
 )
 COSMOS_NANO_OBJECT_SET_METRICS_PATH = (
     MODEL_OUTPUT_TASK_PROBE_DIR / "object_set_forecast/cosmos3_nano_future_window/metrics.json"
 )
+COSMOS_NANO_ACTION_OBJECT_METRICS_PATH = (
+    MODEL_OUTPUT_TASK_PROBE_DIR / "action_object_relation/cosmos3_nano_future_window/metrics.json"
+)
 COSMOS_NANO_TIME_TO_TRANSITION_METRICS_PATH = (
     MODEL_OUTPUT_TASK_PROBE_DIR / "time_to_transition/cosmos3_nano_future_window/metrics.json"
 )
     "action_object_relation": {
         "qwen3_omni_v6_lora": "action_object_relation_macro_f1",
         "cosmos3_super_reasoner": "action_object_relation_macro_f1",
+        "cosmos3_nano_future_window": "action_object_relation_macro_f1",
+    },
+    "caption_grounding": {
+        "cosmos3_super_reasoner": "caption_grounding_iou",
     },
     "long_horizon_next_action": {
         "cosmos3_super_reasoner": "long_horizon_next_action_macro_f1",
 FOUNDATION_METRIC_SOURCE_OVERRIDES = {
     ("qwen3_omni_v6_lora", "action_object_relation"): QWEN_ACTION_OBJECT_METRICS_PATH,
     ("cosmos3_super_reasoner", "action_object_relation"): COSMOS_SUPER_ACTION_OBJECT_METRICS_PATH,
+    ("cosmos3_super_reasoner", "caption_grounding"): COSMOS_SUPER_CAPTION_GROUNDING_METRICS_PATH,
     ("qwen3_omni_v6_lora", "caption_grounding"): QWEN_FUTURE_TASK_METRIC_PATHS["caption_grounding"],
     ("qwen3_omni_v6_lora", "cross_modal_retrieval"): QWEN_FUTURE_TASK_METRIC_PATHS["cross_modal_retrieval"],
     ("qwen3_omni_v6_lora", "temporal_order"): QWEN_FUTURE_TASK_METRIC_PATHS["temporal_order"],
     ("cosmos3_nano_future_window", "long_horizon_next_action"): COSMOS_NANO_LONG_HORIZON_METRICS_PATH,
     ("cosmos3_nano_future_window", "next_subtask_forecast"): COSMOS_NANO_NEXT_SUBTASK_METRICS_PATH,
     ("cosmos3_nano_future_window", "modality_reconstruction"): COSMOS_NANO_MODALITY_RECONSTRUCTION_METRICS_PATH,
+    ("cosmos3_nano_future_window", "action_object_relation"): COSMOS_NANO_ACTION_OBJECT_METRICS_PATH,
     ("cosmos3_nano_future_window", "object_set_forecast"): COSMOS_NANO_OBJECT_SET_METRICS_PATH,
     ("cosmos3_nano_future_window", "time_to_transition"): COSMOS_NANO_TIME_TO_TRANSITION_METRICS_PATH,
     ("cosmos3_super_reasoner", "long_horizon_next_action"): COSMOS_SUPER_LONG_HORIZON_METRICS_PATH,
     "raw128_simple": "128-episode 4430-dim sensor NPZ simple heads; tasks 15/19 use compact proxies.",
     "raw128_neural_mlp": "128-episode 4430-dim sensor NPZ MLP heads; tasks 15/19 use compact proxies.",
     "qwen3_omni_v6_lora": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
+    "cosmos3_super_reasoner": "Verified Cosmos3-Super base-weight Reasoner JSON-task evaluation, plus task 8/16 and a derived task-20 action-boundary timing probe scored from existing verified JSON.",
+    "cosmos3_nano_future_window": "Verified Cosmos3-Nano future-window compatibility metrics, plus tasks 10/13/14/16/17 and a derived task-20 boundary timing probe scored from existing held-out future-window artifacts.",
 }
 PROXY_TASK_IDS = {"interaction_text_prediction", "camera_view_sync_retrieval"}
     cosmos_fd = read_json(COSMOS_SUPER_FD_METRICS_PATH)
     qwen.update(read_json(QWEN_ACTION_OBJECT_METRICS_PATH))
     cosmos_super.update(read_json(COSMOS_SUPER_ACTION_OBJECT_METRICS_PATH))
+    cosmos_super.update(read_json(COSMOS_SUPER_CAPTION_GROUNDING_METRICS_PATH))
     cosmos_super.update(read_json(COSMOS_SUPER_LONG_HORIZON_METRICS_PATH))
     cosmos_super.update(read_json(COSMOS_SUPER_TIME_TO_TRANSITION_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_LONG_HORIZON_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_NEXT_SUBTASK_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_MODALITY_RECONSTRUCTION_METRICS_PATH))
+    cosmos_nano.update(read_json(COSMOS_NANO_ACTION_OBJECT_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_OBJECT_SET_METRICS_PATH))
     cosmos_nano.update(read_json(COSMOS_NANO_TIME_TO_TRANSITION_METRICS_PATH))
     foundation_task_metrics = foundation_task_metric_mapping(qwen)

scripts/omni/score_existing_model_output_task_probes.py CHANGED Viewed

@@ -67,7 +67,6 @@ MODEL_SPECS = {
             "results/omni_finetune/model_output_task_probes_20260616/"
             "cosmos3_nano_future_window_target_map.jsonl"
         ),
-        "action_object_relation_unsupported_reason": "verified future-window predictions do not contain object-set fields",
     },
 }
@@ -661,6 +660,263 @@ def score_cosmos_nano_object_set_from_target_map(
     return metrics
 def score_modality_reconstruction_from_feature_error(
     *,
     model_id: str,
@@ -1142,6 +1398,7 @@ def build_report(summary: dict[str, Any]) -> str:
         task16 = task_results.get("action_object_relation", {})
         task17 = task_results.get("object_set_forecast", {})
         task20 = task_results.get("time_to_transition", {})
         rows.append(
             "| "
             + " | ".join(
@@ -1175,6 +1432,11 @@ def build_report(summary: dict[str, Any]) -> str:
                         if task20.get("time_to_transition_mae") is not None
                         else "n/a"
                     ),
                     result.get("reason") or result.get("source_prediction_jsonl", ""),
                 ]
             )
@@ -1188,8 +1450,8 @@ This package scores only task targets already present in verified held-out
 prediction JSON. It does not run new inference and does not infer targets that
 are absent from a model branch.
-| Method | ID | Status | Scored tasks | Task 13 macro-F1 | Task 14 macro-F1 | Task 16 macro-F1 | Task 17 micro-F1 | Task 20 MAE | Evidence |
-| --- | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | --- |
 {chr(10).join(rows)}
 """
@@ -1213,9 +1475,7 @@ def main() -> int:
             continue
         task_results: dict[str, Any] = {}
         unsupported: dict[str, str] = {}
-        if spec.get("action_object_relation_unsupported_reason"):
-            unsupported["action_object_relation"] = spec["action_object_relation_unsupported_reason"]
-        else:
             metrics = score_action_object_relation(
                 model_id=model_id,
                 spec=spec,
@@ -1230,6 +1490,21 @@ def main() -> int:
                 "action_object_relation_macro_f1": metrics["action_object_relation_macro_f1"],
                 "action_object_relation_accuracy": metrics["action_object_relation_accuracy"],
             }
         if model_id == "cosmos3_nano_future_window":
             manifest_path = workspace / spec["dataset_manifest"]
             metrics = score_cosmos_nano_long_horizon_next_action(
@@ -1278,6 +1553,20 @@ def main() -> int:
                     "object_set_forecast_precision": metrics["object_set_forecast_precision"],
                     "object_set_forecast_recall": metrics["object_set_forecast_recall"],
                 }
             metrics_path = workspace / spec["metrics_json"]
             metrics = score_modality_reconstruction_from_feature_error(
                 model_id=model_id,
@@ -1356,6 +1645,7 @@ def main() -> int:
         ),
         "task_ids_added_to_matrix": [
             "action_object_relation",
             "long_horizon_next_action",
             "modality_reconstruction",
             "next_subtask_forecast",

             "results/omni_finetune/model_output_task_probes_20260616/"
             "cosmos3_nano_future_window_target_map.jsonl"
         ),
     },
 }
     return metrics
+def score_cosmos_nano_action_object_relation_from_target_map(
+    *,
+    model_id: str,
+    spec: dict[str, Any],
+    prediction_jsonl: Path,
+    target_map_jsonl: Path,
+    output_dir: Path,
+    workspace: Path,
+) -> dict[str, Any]:
+    source_rows = read_jsonl(prediction_jsonl)
+    target_by_id = load_target_map(target_map_jsonl)
+    scored_rows: list[dict[str, Any]] = []
+    y_true: list[str] = []
+    y_pred: list[str] = []
+    missing_true_target_count = 0
+    missing_pred_target_count = 0
+    for row in source_rows:
+        future_id = str(row.get("future_record_id") or "")
+        pred_future_id = str(row.get("pred_future_record_id") or "")
+        true_target = target_by_id.get(future_id)
+        if not true_target:
+            missing_true_target_count += 1
+            continue
+        pred_target = target_by_id.get(pred_future_id)
+        if not pred_target:
+            missing_pred_target_count += 1
+        true_relation = relation_label(true_target.get("action"), true_target.get("objects"))
+        if true_relation is None:
+            continue
+        pred_relation = relation_label(
+            pred_target.get("action") if pred_target else None,
+            pred_target.get("objects") if pred_target else None,
+            missing_label=MISSING_PRED_RELATION,
+        )
+        y_true.append(true_relation)
+        y_pred.append(pred_relation or MISSING_PRED_RELATION)
+        scored_rows.append(
+            {
+                "id": row.get("id"),
+                "split": row.get("split"),
+                "episode_id": row.get("episode_id"),
+                "future_record_id": future_id,
+                "pred_future_record_id": pred_future_id,
+                "true_relation": true_relation,
+                "pred_relation": pred_relation,
+                "correct": int(true_relation == pred_relation),
+                "rank": row.get("rank"),
+                "top_k_hit": row.get("top_k_hit"),
+            }
+        )
+    if not y_true:
+        raise RuntimeError(f"no joined future action-object relation targets found in {prediction_jsonl}")
+    label_options = sorted(set(y_true))
+    metrics, per_class, _ = class_metrics(y_true, y_pred, label_options)
+    metrics.update(
+        {
+            "title": f"{spec['label']} Future Action-Object Relation Probe",
+            "status": "pass",
+            "generated_at_utc": datetime.now(timezone.utc).isoformat(timespec="seconds"),
+            "model_id": model_id,
+            "model_label": spec["label"],
+            "task_id": "action_object_relation",
+            "task_number": 16,
+            "task_label": "Action-Object Relation",
+            "metric_key": "action_object_relation_macro_f1",
+            "primary_metric": "action_object_relation_macro_f1",
+            "primary_score": metrics["macro_f1"],
+            "action_object_relation_macro_f1": metrics["macro_f1"],
+            "action_object_relation_accuracy": metrics["accuracy"],
+            "source_prediction_jsonl": relpath(prediction_jsonl, workspace),
+            "source_target_map_jsonl": relpath(target_map_jsonl, workspace),
+            "scope": "held_out_test_existing_future_window_retrieval_probe",
+            "score_policy": (
+                "Derived from existing verified held-out Cosmos3-Nano future-window predictions. "
+                "The true future record and retrieved future record are joined to the public-safe "
+                "target map, then scored on the action plus object-set relation."
+            ),
+            "normalization_policy": (
+                "Action text is whitespace-normalized. Objects are casefolded, deduplicated, "
+                "sorted, and joined into a canonical relation label before macro-F1 scoring."
+            ),
+            "known_limitation": (
+                "This is a retrieval-derived future relation score, not a separately prompted "
+                "action-object generation head."
+            ),
+            "total_prediction_rows": len(source_rows),
+            "target_map_rows": len(target_by_id),
+            "scored_rows": len(scored_rows),
+            "missing_true_target_count": missing_true_target_count,
+            "missing_pred_target_count": missing_pred_target_count,
+            "artifact_files": {
+                "metrics_json": relpath(output_dir / "metrics.json", workspace),
+                "predictions_csv": relpath(output_dir / "predictions.csv", workspace),
+                "per_class_metrics_csv": relpath(output_dir / "per_class_metrics.csv", workspace),
+            },
+        }
+    )
+    write_json(output_dir / "metrics.json", metrics)
+    write_csv(
+        output_dir / "predictions.csv",
+        scored_rows,
+        [
+            "id",
+            "split",
+            "episode_id",
+            "future_record_id",
+            "pred_future_record_id",
+            "true_relation",
+            "pred_relation",
+            "correct",
+            "rank",
+            "top_k_hit",
+        ],
+    )
+    write_csv(
+        output_dir / "per_class_metrics.csv",
+        per_class,
+        ["class_name", "support", "predicted", "precision", "recall", "f1"],
+    )
+    return metrics
+def parse_window(value: Any) -> tuple[float, float] | None:
+    if not isinstance(value, dict):
+        return None
+    try:
+        start = float(value.get("start_frame"))
+        end = float(value.get("end_frame"))
+    except (TypeError, ValueError):
+        return None
+    if end < start:
+        return None
+    return start, end
+def interval_iou(a: tuple[float, float], b: tuple[float, float]) -> float:
+    inter = max(0.0, min(a[1], b[1]) - max(a[0], b[0]) + 1.0)
+    union = max(a[1], b[1]) - min(a[0], b[0]) + 1.0
+    return inter / union if union > 0 else 0.0
+def score_caption_grounding_from_evidence_window(
+    *,
+    model_id: str,
+    spec: dict[str, Any],
+    prediction_jsonl: Path,
+    output_dir: Path,
+    workspace: Path,
+) -> dict[str, Any]:
+    source_rows = read_jsonl(prediction_jsonl)
+    scored_rows: list[dict[str, Any]] = []
+    ious: list[float] = []
+    center_hits = 0
+    within_20_hits = 0
+    missing_pred_count = 0
+    for row in source_rows:
+        true_json = row.get("true_json") if isinstance(row.get("true_json"), dict) else {}
+        pred_json = row.get("pred_json") if isinstance(row.get("pred_json"), dict) else {}
+        true_window = parse_window(true_json.get("evidence_window"))
+        if true_window is None:
+            continue
+        pred_window = parse_window(pred_json.get("evidence_window"))
+        if pred_window is None:
+            missing_pred_count += 1
+            iou = 0.0
+            pred_center = None
+            center_error = None
+        else:
+            iou = interval_iou(true_window, pred_window)
+            pred_center = (pred_window[0] + pred_window[1]) / 2.0
+            true_center = (true_window[0] + true_window[1]) / 2.0
+            center_error = abs(pred_center - true_center)
+            center_hits += int(true_window[0] <= pred_center <= true_window[1])
+            within_20_hits += int(center_error <= 20.0)
+        ious.append(iou)
+        scored_rows.append(
+            {
+                "id": row.get("id"),
+                "split": row.get("split"),
+                "episode_id": row.get("episode_id"),
+                "true_start_frame": true_window[0],
+                "true_end_frame": true_window[1],
+                "pred_start_frame": pred_window[0] if pred_window else None,
+                "pred_end_frame": pred_window[1] if pred_window else None,
+                "pred_center_frame": pred_center,
+                "center_error_frames": center_error,
+                "iou": iou,
+            }
+        )
+    if not scored_rows:
+        raise RuntimeError(f"no evidence-window targets found in {prediction_jsonl}")
+    mean_iou = sum(ious) / len(ious)
+    center_hit_rate = center_hits / len(scored_rows)
+    within_20_rate = within_20_hits / len(scored_rows)
+    metrics = {
+        "title": f"{spec['label']} Evidence-Window Grounding Probe",
+        "status": "pass",
+        "generated_at_utc": datetime.now(timezone.utc).isoformat(timespec="seconds"),
+        "model_id": model_id,
+        "model_label": spec["label"],
+        "task_id": "caption_grounding",
+        "task_number": 8,
+        "task_label": "Language Grounding",
+        "metric_key": "caption_grounding_iou",
+        "primary_metric": "caption_grounding_iou",
+        "primary_score": mean_iou,
+        "caption_grounding_iou": mean_iou,
+        "caption_grounding_center_hit_rate": center_hit_rate,
+        "caption_grounding_within_20_frames": within_20_rate,
+        "source_prediction_jsonl": relpath(prediction_jsonl, workspace),
+        "scope": "held_out_test_existing_verified_prediction_json",
+        "score_policy": (
+            "Derived from existing verified held-out structured JSON predictions. The task target is "
+            "the evidence_window field already present in the true JSON; missing or invalid predicted "
+            "evidence windows receive zero IoU."
+        ),
+        "normalization_policy": (
+            "Evidence windows are frame intervals. The primary score is mean interval IoU; center-hit "
+            "and within-20-frame rates are reported as diagnostics."
+        ),
+        "known_limitation": (
+            "This is an evidence-window localization score, not a candidate-set retrieval MRR."
+        ),
+        "total_prediction_rows": len(source_rows),
+        "scored_rows": len(scored_rows),
+        "missing_pred_evidence_window_count": missing_pred_count,
+        "artifact_files": {
+            "metrics_json": relpath(output_dir / "metrics.json", workspace),
+            "predictions_csv": relpath(output_dir / "predictions.csv", workspace),
+        },
+    }
+    write_json(output_dir / "metrics.json", metrics)
+    write_csv(
+        output_dir / "predictions.csv",
+        scored_rows,
+        [
+            "id",
+            "split",
+            "episode_id",
+            "true_start_frame",
+            "true_end_frame",
+            "pred_start_frame",
+            "pred_end_frame",
+            "pred_center_frame",
+            "center_error_frames",
+            "iou",
+        ],
+    )
+    return metrics
 def score_modality_reconstruction_from_feature_error(
     *,
     model_id: str,
         task16 = task_results.get("action_object_relation", {})
         task17 = task_results.get("object_set_forecast", {})
         task20 = task_results.get("time_to_transition", {})
+        task8 = task_results.get("caption_grounding", {})
         rows.append(
             "| "
             + " | ".join(
                         if task20.get("time_to_transition_mae") is not None
                         else "n/a"
                     ),
+                    (
+                        f"{task8.get('caption_grounding_iou', 0.0):.6f}"
+                        if task8.get("caption_grounding_iou") is not None
+                        else "n/a"
+                    ),
                     result.get("reason") or result.get("source_prediction_jsonl", ""),
                 ]
             )
 prediction JSON. It does not run new inference and does not infer targets that
 are absent from a model branch.
+| Method | ID | Status | Scored tasks | Task 13 macro-F1 | Task 14 macro-F1 | Task 16 macro-F1 | Task 17 micro-F1 | Task 20 MAE | Task 8 IoU | Evidence |
+| --- | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | --- |
 {chr(10).join(rows)}
 """
             continue
         task_results: dict[str, Any] = {}
         unsupported: dict[str, str] = {}
+        if model_id != "cosmos3_nano_future_window":
             metrics = score_action_object_relation(
                 model_id=model_id,
                 spec=spec,
                 "action_object_relation_macro_f1": metrics["action_object_relation_macro_f1"],
                 "action_object_relation_accuracy": metrics["action_object_relation_accuracy"],
             }
+        if model_id == "cosmos3_super_reasoner":
+            metrics = score_caption_grounding_from_evidence_window(
+                model_id=model_id,
+                spec=spec,
+                prediction_jsonl=prediction_path,
+                output_dir=output_dir / "caption_grounding" / model_id,
+                workspace=workspace,
+            )
+            task_results["caption_grounding"] = {
+                "source_metrics_json": metrics["artifact_files"]["metrics_json"],
+                "scored_rows": metrics["scored_rows"],
+                "caption_grounding_iou": metrics["caption_grounding_iou"],
+                "caption_grounding_center_hit_rate": metrics["caption_grounding_center_hit_rate"],
+                "missing_pred_evidence_window_count": metrics["missing_pred_evidence_window_count"],
+            }
         if model_id == "cosmos3_nano_future_window":
             manifest_path = workspace / spec["dataset_manifest"]
             metrics = score_cosmos_nano_long_horizon_next_action(
                     "object_set_forecast_precision": metrics["object_set_forecast_precision"],
                     "object_set_forecast_recall": metrics["object_set_forecast_recall"],
                 }
+                metrics = score_cosmos_nano_action_object_relation_from_target_map(
+                    model_id=model_id,
+                    spec=spec,
+                    prediction_jsonl=prediction_path,
+                    target_map_jsonl=target_map_path,
+                    output_dir=output_dir / "action_object_relation" / model_id,
+                    workspace=workspace,
+                )
+                task_results["action_object_relation"] = {
+                    "source_metrics_json": metrics["artifact_files"]["metrics_json"],
+                    "scored_rows": metrics["scored_rows"],
+                    "action_object_relation_macro_f1": metrics["action_object_relation_macro_f1"],
+                    "action_object_relation_accuracy": metrics["action_object_relation_accuracy"],
+                }
             metrics_path = workspace / spec["metrics_json"]
             metrics = score_modality_reconstruction_from_feature_error(
                 model_id=model_id,
         ),
         "task_ids_added_to_matrix": [
             "action_object_relation",
+            "caption_grounding",
             "long_horizon_next_action",
             "modality_reconstruction",
             "next_subtask_forecast",