Add files using upload-large-folder tool
Browse files- TASK_METHOD_20_GAP_AUDIT.md +5 -8
- TASK_METHOD_20_RESULT_MATRIX.md +4 -4
- assets/charts/episode128_task_model_radar.svg +6 -3
- assets/charts/unified_task_model_radar.svg +5 -2
- data/episode128_task_model_radar.json +60 -60
- data/publication_audit.json +3 -3
- data/single_episode_task_model_radar.json +1 -1
- data/task_method_20_gap_audit.json +13 -55
- data/task_method_20_result_matrix.json +36 -36
- data/task_surface_integrity.json +1 -1
- data/unified_task_model_radar.json +70 -70
- data/website_integrity.json +7 -7
- docs/data/episode128_task_model_radar.json +60 -60
- docs/data/publication_audit.json +3 -3
- docs/data/single_episode_task_model_radar.json +1 -1
- docs/data/task_method_20_gap_audit.json +13 -55
- docs/data/task_method_20_result_matrix.json +36 -36
- docs/data/task_surface_integrity.json +1 -1
- docs/data/unified_task_model_radar.json +70 -70
- docs/data/website_integrity.json +7 -7
- metrics/episode128_task_model_radar.json +60 -60
- metrics/publication_audit.json +3 -3
- metrics/single_episode_task_model_radar.json +1 -1
- metrics/task_method_20_gap_audit.json +13 -55
- metrics/task_method_20_result_matrix.json +36 -36
- metrics/task_surface_integrity.json +1 -1
- metrics/unified_task_model_radar.json +70 -70
- metrics/website_integrity.json +7 -7
TASK_METHOD_20_GAP_AUDIT.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
# Task Method 20-Result Gap Audit
|
| 2 |
|
| 3 |
-
Generated: `2026-06-
|
| 4 |
|
| 5 |
This audit is the explicit gap ledger for the 9-method x 20-task result matrix.
|
| 6 |
It keeps missing cells visible while preserving the rule that a numeric score
|
|
@@ -9,8 +9,8 @@ requires a real task target and source artifact.
|
|
| 9 |
## Score Summary
|
| 10 |
|
| 11 |
- Method-task records: `180`
|
| 12 |
-
- Numeric scored records: `
|
| 13 |
-
- Scoreless records: `
|
| 14 |
- Proxy-scored records: `4`
|
| 15 |
- Source matrix: [`docs/data/task_method_20_result_matrix.json`](docs/data/task_method_20_result_matrix.json)
|
| 16 |
|
|
@@ -24,7 +24,7 @@ requires a real task target and source artifact.
|
|
| 24 |
| 128ep Metadata NN | metadata128_neural_mlp | 6/20 | 14 | 0 | not_supported_by_metadata_only_package: 14, scored: 6 |
|
| 25 |
| 128ep Raw Simple | raw128_simple | 20/20 | 0 | 2 | proxy_scored: 2, scored: 18 |
|
| 26 |
| 128ep Raw NN | raw128_neural_mlp | 20/20 | 0 | 2 | proxy_scored: 2, scored: 18 |
|
| 27 |
-
| Qwen3-Omni v6 LoRA | qwen3_omni_v6_lora |
|
| 28 |
| Cosmos3-Super Reasoner | cosmos3_super_reasoner | 7/20 | 13 | 0 | not_evaluated_in_verified_package: 13, scored: 7 |
|
| 29 |
| Cosmos3-Nano Future Window | cosmos3_nano_future_window | 5/20 | 15 | 0 | not_evaluated_in_verified_package: 15, scored: 5 |
|
| 30 |
|
|
@@ -32,7 +32,7 @@ requires a real task target and source artifact.
|
|
| 32 |
|
| 33 |
| Status | Count | Next step |
|
| 34 |
| --- | --- | --- |
|
| 35 |
-
| not_evaluated_in_verified_package |
|
| 36 |
| not_supported_by_metadata_only_package | 22 | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
| 37 |
| unsupported_without_required_target | 4 | Export the missing target field for this 128-episode method, then rerun the same train/validation/test split. |
|
| 38 |
|
|
@@ -61,12 +61,10 @@ requires a real task target and source artifact.
|
|
| 61 |
| 10 | Cross-Modal Reconstruction | Cosmos3-Super Reasoner | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 62 |
| 10 | Cross-Modal Reconstruction | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 63 |
| 11 | Temporal Order Verification | 128ep Metadata NN | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
| 64 |
-
| 11 | Temporal Order Verification | Qwen3-Omni v6 LoRA | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 65 |
| 11 | Temporal Order Verification | Cosmos3-Super Reasoner | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 66 |
| 11 | Temporal Order Verification | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 67 |
| 12 | Multimodal Synchronization Detection | 128ep Metadata Simple | unsupported | Export the missing target field for this 128-episode method, then rerun the same train/validation/test split. |
|
| 68 |
| 12 | Multimodal Synchronization Detection | 128ep Metadata NN | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
| 69 |
-
| 12 | Multimodal Synchronization Detection | Qwen3-Omni v6 LoRA | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 70 |
| 12 | Multimodal Synchronization Detection | Cosmos3-Super Reasoner | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 71 |
| 12 | Multimodal Synchronization Detection | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 72 |
| 13 | Long-Horizon Next-Action Forecasting | 128ep Metadata Simple | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
|
@@ -101,7 +99,6 @@ requires a real task target and source artifact.
|
|
| 101 |
| 19 | Camera-View Synchronization Retrieval | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 102 |
| 20 | Time-to-Next-Transition Regression | 128ep Metadata Simple | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
| 103 |
| 20 | Time-to-Next-Transition Regression | 128ep Metadata NN | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
| 104 |
-
| 20 | Time-to-Next-Transition Regression | Qwen3-Omni v6 LoRA | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 105 |
| 20 | Time-to-Next-Transition Regression | Cosmos3-Super Reasoner | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 106 |
| 20 | Time-to-Next-Transition Regression | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 107 |
|
|
|
|
| 1 |
# Task Method 20-Result Gap Audit
|
| 2 |
|
| 3 |
+
Generated: `2026-06-17T21:17:51+00:00`
|
| 4 |
|
| 5 |
This audit is the explicit gap ledger for the 9-method x 20-task result matrix.
|
| 6 |
It keeps missing cells visible while preserving the rule that a numeric score
|
|
|
|
| 9 |
## Score Summary
|
| 10 |
|
| 11 |
- Method-task records: `180`
|
| 12 |
+
- Numeric scored records: `119`
|
| 13 |
+
- Scoreless records: `61`
|
| 14 |
- Proxy-scored records: `4`
|
| 15 |
- Source matrix: [`docs/data/task_method_20_result_matrix.json`](docs/data/task_method_20_result_matrix.json)
|
| 16 |
|
|
|
|
| 24 |
| 128ep Metadata NN | metadata128_neural_mlp | 6/20 | 14 | 0 | not_supported_by_metadata_only_package: 14, scored: 6 |
|
| 25 |
| 128ep Raw Simple | raw128_simple | 20/20 | 0 | 2 | proxy_scored: 2, scored: 18 |
|
| 26 |
| 128ep Raw NN | raw128_neural_mlp | 20/20 | 0 | 2 | proxy_scored: 2, scored: 18 |
|
| 27 |
+
| Qwen3-Omni v6 LoRA | qwen3_omni_v6_lora | 13/20 | 7 | 0 | not_evaluated_in_verified_package: 7, scored: 13 |
|
| 28 |
| Cosmos3-Super Reasoner | cosmos3_super_reasoner | 7/20 | 13 | 0 | not_evaluated_in_verified_package: 13, scored: 7 |
|
| 29 |
| Cosmos3-Nano Future Window | cosmos3_nano_future_window | 5/20 | 15 | 0 | not_evaluated_in_verified_package: 15, scored: 5 |
|
| 30 |
|
|
|
|
| 32 |
|
| 33 |
| Status | Count | Next step |
|
| 34 |
| --- | --- | --- |
|
| 35 |
+
| not_evaluated_in_verified_package | 35 | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 36 |
| not_supported_by_metadata_only_package | 22 | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
| 37 |
| unsupported_without_required_target | 4 | Export the missing target field for this 128-episode method, then rerun the same train/validation/test split. |
|
| 38 |
|
|
|
|
| 61 |
| 10 | Cross-Modal Reconstruction | Cosmos3-Super Reasoner | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 62 |
| 10 | Cross-Modal Reconstruction | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 63 |
| 11 | Temporal Order Verification | 128ep Metadata NN | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
|
|
|
| 64 |
| 11 | Temporal Order Verification | Cosmos3-Super Reasoner | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 65 |
| 11 | Temporal Order Verification | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 66 |
| 12 | Multimodal Synchronization Detection | 128ep Metadata Simple | unsupported | Export the missing target field for this 128-episode method, then rerun the same train/validation/test split. |
|
| 67 |
| 12 | Multimodal Synchronization Detection | 128ep Metadata NN | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
|
|
|
| 68 |
| 12 | Multimodal Synchronization Detection | Cosmos3-Super Reasoner | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 69 |
| 12 | Multimodal Synchronization Detection | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 70 |
| 13 | Long-Horizon Next-Action Forecasting | 128ep Metadata Simple | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
|
|
|
| 99 |
| 19 | Camera-View Synchronization Retrieval | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 100 |
| 20 | Time-to-Next-Transition Regression | 128ep Metadata Simple | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
| 101 |
| 20 | Time-to-Next-Transition Regression | 128ep Metadata NN | not supported | Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score. |
|
|
|
|
| 102 |
| 20 | Time-to-Next-Transition Regression | Cosmos3-Super Reasoner | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 103 |
| 20 | Time-to-Next-Transition Regression | Cosmos3-Nano Future Window | not evaluated | Generate verified model outputs for this task contract and score them against the held-out labels. |
|
| 104 |
|
TASK_METHOD_20_RESULT_MATRIX.md
CHANGED
|
@@ -12,7 +12,7 @@ Legend: `score` = numeric task score, `proxy` = documented raw128 compact proxy
|
|
| 12 |
| 128ep Metadata NN | 20 | 6 | 0 | 14 | not supported 14, scored 6 |
|
| 13 |
| 128ep Raw Simple | 20 | 20 | 2 | 0 | proxy scored 2, scored 18 |
|
| 14 |
| 128ep Raw NN | 20 | 20 | 2 | 0 | proxy scored 2, scored 18 |
|
| 15 |
-
| Qwen3-Omni v6 LoRA | 20 |
|
| 16 |
| Cosmos3-Super Reasoner | 20 | 7 | 0 | 13 | not evaluated 13, scored 7 |
|
| 17 |
| Cosmos3-Nano Future Window | 20 | 5 | 0 | 15 | not evaluated 15, scored 5 |
|
| 18 |
|
|
@@ -28,8 +28,8 @@ Legend: `score` = numeric task score, `proxy` = documented raw128 compact proxy
|
|
| 28 |
| 08 | Language Grounding | score | score | score | not supported | score | score | not evaluated | not evaluated | not evaluated |
|
| 29 |
| 09 | Cross-Modal Retrieval | score | score | unsupported | not supported | score | score | not evaluated | not evaluated | score |
|
| 30 |
| 10 | Cross-Modal Reconstruction | score | score | unsupported | not supported | score | score | not evaluated | not evaluated | not evaluated |
|
| 31 |
-
| 11 | Temporal Order Verification | score | score | score | not supported | score | score |
|
| 32 |
-
| 12 | Multimodal Synchronization Detection | score | score | unsupported | not supported | score | score |
|
| 33 |
| 13 | Long-Horizon Next-Action Forecasting | score | score | not supported | not supported | score | score | score | not evaluated | not evaluated |
|
| 34 |
| 14 | Long-Horizon Next-Subtask Forecasting | score | score | not supported | not supported | score | score | score | not evaluated | not evaluated |
|
| 35 |
| 15 | Interaction Text Prediction | score | score | not supported | not supported | proxy | proxy | not evaluated | not evaluated | not evaluated |
|
|
@@ -37,6 +37,6 @@ Legend: `score` = numeric task score, `proxy` = documented raw128 compact proxy
|
|
| 37 |
| 17 | Future Object-Set Forecasting | score | score | not supported | not supported | score | score | score | not evaluated | not evaluated |
|
| 38 |
| 18 | IMU-to-Hand Pose Reconstruction | score | score | not supported | not supported | score | score | not evaluated | not evaluated | not evaluated |
|
| 39 |
| 19 | Camera-View Synchronization Retrieval | score | score | not supported | not supported | proxy | proxy | not evaluated | not evaluated | not evaluated |
|
| 40 |
-
| 20 | Time-to-Next-Transition Regression | score | score | not supported | not supported | score | score |
|
| 41 |
|
| 42 |
Sources and raw values are in `docs/data/task_method_20_result_matrix.json` and `docs/data/unified_task_model_radar.json`.
|
|
|
|
| 12 |
| 128ep Metadata NN | 20 | 6 | 0 | 14 | not supported 14, scored 6 |
|
| 13 |
| 128ep Raw Simple | 20 | 20 | 2 | 0 | proxy scored 2, scored 18 |
|
| 14 |
| 128ep Raw NN | 20 | 20 | 2 | 0 | proxy scored 2, scored 18 |
|
| 15 |
+
| Qwen3-Omni v6 LoRA | 20 | 13 | 0 | 7 | not evaluated 7, scored 13 |
|
| 16 |
| Cosmos3-Super Reasoner | 20 | 7 | 0 | 13 | not evaluated 13, scored 7 |
|
| 17 |
| Cosmos3-Nano Future Window | 20 | 5 | 0 | 15 | not evaluated 15, scored 5 |
|
| 18 |
|
|
|
|
| 28 |
| 08 | Language Grounding | score | score | score | not supported | score | score | not evaluated | not evaluated | not evaluated |
|
| 29 |
| 09 | Cross-Modal Retrieval | score | score | unsupported | not supported | score | score | not evaluated | not evaluated | score |
|
| 30 |
| 10 | Cross-Modal Reconstruction | score | score | unsupported | not supported | score | score | not evaluated | not evaluated | not evaluated |
|
| 31 |
+
| 11 | Temporal Order Verification | score | score | score | not supported | score | score | score | not evaluated | not evaluated |
|
| 32 |
+
| 12 | Multimodal Synchronization Detection | score | score | unsupported | not supported | score | score | score | not evaluated | not evaluated |
|
| 33 |
| 13 | Long-Horizon Next-Action Forecasting | score | score | not supported | not supported | score | score | score | not evaluated | not evaluated |
|
| 34 |
| 14 | Long-Horizon Next-Subtask Forecasting | score | score | not supported | not supported | score | score | score | not evaluated | not evaluated |
|
| 35 |
| 15 | Interaction Text Prediction | score | score | not supported | not supported | proxy | proxy | not evaluated | not evaluated | not evaluated |
|
|
|
|
| 37 |
| 17 | Future Object-Set Forecasting | score | score | not supported | not supported | score | score | score | not evaluated | not evaluated |
|
| 38 |
| 18 | IMU-to-Hand Pose Reconstruction | score | score | not supported | not supported | score | score | not evaluated | not evaluated | not evaluated |
|
| 39 |
| 19 | Camera-View Synchronization Retrieval | score | score | not supported | not supported | proxy | proxy | not evaluated | not evaluated | not evaluated |
|
| 40 |
+
| 20 | Time-to-Next-Transition Regression | score | score | not supported | not supported | score | score | score | not evaluated | not evaluated |
|
| 41 |
|
| 42 |
Sources and raw values are in `docs/data/task_method_20_result_matrix.json` and `docs/data/unified_task_model_radar.json`.
|
assets/charts/episode128_task_model_radar.svg
CHANGED
|
|
assets/charts/unified_task_model_radar.svg
CHANGED
|
|
data/episode128_task_model_radar.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"title": "128-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"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.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 7,
|
| 8 |
"method_task_record_count": 140,
|
| 9 |
-
"scored_method_task_count":
|
| 10 |
"normalization_policy": {
|
| 11 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 12 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
@@ -127,17 +127,17 @@
|
|
| 127 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 128 |
"plotted_as": "colored point overlay",
|
| 129 |
"result_record_count": 20,
|
| 130 |
-
"scored_task_count":
|
| 131 |
-
"covered_task_count":
|
| 132 |
"proxy_scored_task_count": 0,
|
| 133 |
-
"scoreless_task_count":
|
| 134 |
"unsupported_task_count": 0,
|
| 135 |
-
"not_evaluated_task_count":
|
| 136 |
"status_counts": {
|
| 137 |
-
"not_evaluated_in_verified_package":
|
| 138 |
-
"scored":
|
| 139 |
},
|
| 140 |
-
"coverage_fraction": 0.
|
| 141 |
"result_record_fraction": 1.0
|
| 142 |
},
|
| 143 |
{
|
|
@@ -1157,15 +1157,15 @@
|
|
| 1157 |
"status_label": "scored"
|
| 1158 |
},
|
| 1159 |
"qwen3_omni_v6_lora": {
|
| 1160 |
-
"raw":
|
| 1161 |
-
"metric_key": "
|
| 1162 |
-
"source":
|
| 1163 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1164 |
-
"status": "
|
| 1165 |
-
"reason":
|
| 1166 |
-
"normalized_score":
|
| 1167 |
-
"raw_text": "
|
| 1168 |
-
"status_label": "
|
| 1169 |
},
|
| 1170 |
"cosmos3_super_reasoner": {
|
| 1171 |
"raw": null,
|
|
@@ -1248,15 +1248,15 @@
|
|
| 1248 |
"status_label": "scored"
|
| 1249 |
},
|
| 1250 |
"qwen3_omni_v6_lora": {
|
| 1251 |
-
"raw":
|
| 1252 |
-
"metric_key": "
|
| 1253 |
-
"source":
|
| 1254 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1255 |
-
"status": "
|
| 1256 |
-
"reason":
|
| 1257 |
-
"normalized_score":
|
| 1258 |
-
"raw_text": "
|
| 1259 |
-
"status_label": "
|
| 1260 |
},
|
| 1261 |
"cosmos3_super_reasoner": {
|
| 1262 |
"raw": null,
|
|
@@ -1976,15 +1976,15 @@
|
|
| 1976 |
"status_label": "scored"
|
| 1977 |
},
|
| 1978 |
"qwen3_omni_v6_lora": {
|
| 1979 |
-
"raw":
|
| 1980 |
-
"metric_key": "
|
| 1981 |
-
"source":
|
| 1982 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1983 |
-
"status": "
|
| 1984 |
-
"reason":
|
| 1985 |
-
"normalized_score":
|
| 1986 |
-
"raw_text": "
|
| 1987 |
-
"status_label": "
|
| 1988 |
},
|
| 1989 |
"cosmos3_super_reasoner": {
|
| 1990 |
"raw": null,
|
|
@@ -3350,17 +3350,17 @@
|
|
| 3350 |
"task_label": "Temporal Order Verification",
|
| 3351 |
"series_id": "qwen3_omni_v6_lora",
|
| 3352 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3353 |
-
"status": "
|
| 3354 |
-
"status_label": "
|
| 3355 |
-
"scored":
|
| 3356 |
"proxy_scored": false,
|
| 3357 |
-
"raw":
|
| 3358 |
-
"raw_text": "
|
| 3359 |
-
"normalized_score":
|
| 3360 |
-
"metric_key": "
|
| 3361 |
-
"source":
|
| 3362 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3363 |
-
"reason":
|
| 3364 |
},
|
| 3365 |
{
|
| 3366 |
"task_number": 11,
|
|
@@ -3476,17 +3476,17 @@
|
|
| 3476 |
"task_label": "Multimodal Synchronization Detection",
|
| 3477 |
"series_id": "qwen3_omni_v6_lora",
|
| 3478 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3479 |
-
"status": "
|
| 3480 |
-
"status_label": "
|
| 3481 |
-
"scored":
|
| 3482 |
"proxy_scored": false,
|
| 3483 |
-
"raw":
|
| 3484 |
-
"raw_text": "
|
| 3485 |
-
"normalized_score":
|
| 3486 |
-
"metric_key": "
|
| 3487 |
-
"source":
|
| 3488 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3489 |
-
"reason":
|
| 3490 |
},
|
| 3491 |
{
|
| 3492 |
"task_number": 12,
|
|
@@ -4484,17 +4484,17 @@
|
|
| 4484 |
"task_label": "Time-to-Next-Transition Regression",
|
| 4485 |
"series_id": "qwen3_omni_v6_lora",
|
| 4486 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4487 |
-
"status": "
|
| 4488 |
-
"status_label": "
|
| 4489 |
-
"scored":
|
| 4490 |
"proxy_scored": false,
|
| 4491 |
-
"raw":
|
| 4492 |
-
"raw_text": "
|
| 4493 |
-
"normalized_score":
|
| 4494 |
-
"metric_key": "
|
| 4495 |
-
"source":
|
| 4496 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4497 |
-
"reason":
|
| 4498 |
},
|
| 4499 |
{
|
| 4500 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "128-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"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.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 7,
|
| 8 |
"method_task_record_count": 140,
|
| 9 |
+
"scored_method_task_count": 79,
|
| 10 |
"normalization_policy": {
|
| 11 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 12 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
|
|
| 127 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 128 |
"plotted_as": "colored point overlay",
|
| 129 |
"result_record_count": 20,
|
| 130 |
+
"scored_task_count": 13,
|
| 131 |
+
"covered_task_count": 13,
|
| 132 |
"proxy_scored_task_count": 0,
|
| 133 |
+
"scoreless_task_count": 7,
|
| 134 |
"unsupported_task_count": 0,
|
| 135 |
+
"not_evaluated_task_count": 7,
|
| 136 |
"status_counts": {
|
| 137 |
+
"not_evaluated_in_verified_package": 7,
|
| 138 |
+
"scored": 13
|
| 139 |
},
|
| 140 |
+
"coverage_fraction": 0.65,
|
| 141 |
"result_record_fraction": 1.0
|
| 142 |
},
|
| 143 |
{
|
|
|
|
| 1157 |
"status_label": "scored"
|
| 1158 |
},
|
| 1159 |
"qwen3_omni_v6_lora": {
|
| 1160 |
+
"raw": 0.40984631701404173,
|
| 1161 |
+
"metric_key": "temporal_order_f1",
|
| 1162 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1163 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1164 |
+
"status": "scored",
|
| 1165 |
+
"reason": null,
|
| 1166 |
+
"normalized_score": 0.40984631701404173,
|
| 1167 |
+
"raw_text": "0.4098",
|
| 1168 |
+
"status_label": "scored"
|
| 1169 |
},
|
| 1170 |
"cosmos3_super_reasoner": {
|
| 1171 |
"raw": null,
|
|
|
|
| 1248 |
"status_label": "scored"
|
| 1249 |
},
|
| 1250 |
"qwen3_omni_v6_lora": {
|
| 1251 |
+
"raw": 0.3344936184319576,
|
| 1252 |
+
"metric_key": "misalignment_detection_f1",
|
| 1253 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 1254 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1255 |
+
"status": "scored",
|
| 1256 |
+
"reason": null,
|
| 1257 |
+
"normalized_score": 0.3344936184319576,
|
| 1258 |
+
"raw_text": "0.3345",
|
| 1259 |
+
"status_label": "scored"
|
| 1260 |
},
|
| 1261 |
"cosmos3_super_reasoner": {
|
| 1262 |
"raw": null,
|
|
|
|
| 1976 |
"status_label": "scored"
|
| 1977 |
},
|
| 1978 |
"qwen3_omni_v6_lora": {
|
| 1979 |
+
"raw": 134.0687422166874,
|
| 1980 |
+
"metric_key": "time_to_transition_mae",
|
| 1981 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 1982 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1983 |
+
"status": "scored",
|
| 1984 |
+
"reason": null,
|
| 1985 |
+
"normalized_score": 0.07859666766782253,
|
| 1986 |
+
"raw_text": "134.07",
|
| 1987 |
+
"status_label": "scored"
|
| 1988 |
},
|
| 1989 |
"cosmos3_super_reasoner": {
|
| 1990 |
"raw": null,
|
|
|
|
| 3350 |
"task_label": "Temporal Order Verification",
|
| 3351 |
"series_id": "qwen3_omni_v6_lora",
|
| 3352 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3353 |
+
"status": "scored",
|
| 3354 |
+
"status_label": "scored",
|
| 3355 |
+
"scored": true,
|
| 3356 |
"proxy_scored": false,
|
| 3357 |
+
"raw": 0.40984631701404173,
|
| 3358 |
+
"raw_text": "0.4098",
|
| 3359 |
+
"normalized_score": 0.40984631701404173,
|
| 3360 |
+
"metric_key": "temporal_order_f1",
|
| 3361 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 3362 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3363 |
+
"reason": null
|
| 3364 |
},
|
| 3365 |
{
|
| 3366 |
"task_number": 11,
|
|
|
|
| 3476 |
"task_label": "Multimodal Synchronization Detection",
|
| 3477 |
"series_id": "qwen3_omni_v6_lora",
|
| 3478 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3479 |
+
"status": "scored",
|
| 3480 |
+
"status_label": "scored",
|
| 3481 |
+
"scored": true,
|
| 3482 |
"proxy_scored": false,
|
| 3483 |
+
"raw": 0.3344936184319576,
|
| 3484 |
+
"raw_text": "0.3345",
|
| 3485 |
+
"normalized_score": 0.3344936184319576,
|
| 3486 |
+
"metric_key": "misalignment_detection_f1",
|
| 3487 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 3488 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3489 |
+
"reason": null
|
| 3490 |
},
|
| 3491 |
{
|
| 3492 |
"task_number": 12,
|
|
|
|
| 4484 |
"task_label": "Time-to-Next-Transition Regression",
|
| 4485 |
"series_id": "qwen3_omni_v6_lora",
|
| 4486 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4487 |
+
"status": "scored",
|
| 4488 |
+
"status_label": "scored",
|
| 4489 |
+
"scored": true,
|
| 4490 |
"proxy_scored": false,
|
| 4491 |
+
"raw": 134.0687422166874,
|
| 4492 |
+
"raw_text": "134.07",
|
| 4493 |
+
"normalized_score": 0.07859666766782253,
|
| 4494 |
+
"metric_key": "time_to_transition_mae",
|
| 4495 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 4496 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4497 |
+
"reason": null
|
| 4498 |
},
|
| 4499 |
{
|
| 4500 |
"task_number": 20,
|
data/publication_audit.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-17T21:
|
| 4 |
"checks": [
|
| 5 |
{
|
| 6 |
"name": "required_publication_assets_present",
|
|
@@ -206,8 +206,8 @@
|
|
| 206 |
"github_repo": {
|
| 207 |
"root": "repo",
|
| 208 |
"exists": true,
|
| 209 |
-
"file_count":
|
| 210 |
-
"text_file_count":
|
| 211 |
"largest_file": {
|
| 212 |
"path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
|
| 213 |
"bytes": 55702978
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:35+00:00",
|
| 4 |
"checks": [
|
| 5 |
{
|
| 6 |
"name": "required_publication_assets_present",
|
|
|
|
| 206 |
"github_repo": {
|
| 207 |
"root": "repo",
|
| 208 |
"exists": true,
|
| 209 |
+
"file_count": 1250,
|
| 210 |
+
"text_file_count": 1052,
|
| 211 |
"largest_file": {
|
| 212 |
"path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
|
| 213 |
"bytes": 55702978
|
data/single_episode_task_model_radar.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"title": "Single-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"description": "Minimal and Neural MLP baselines on the one public sample episode, both scored on all 20 task contracts.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 2,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Single-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"description": "Minimal and Neural MLP baselines on the one public sample episode, both scored on all 20 task contracts.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 2,
|
data/task_method_20_gap_audit.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"generated_at_utc": "2026-06-
|
| 3 |
"immediate_actions": [
|
| 4 |
{
|
| 5 |
"artifact": "docs/data/task_method_20_gap_audit.json",
|
| 6 |
"id": "gap_audit",
|
| 7 |
-
"purpose": "Keep the
|
| 8 |
},
|
| 9 |
{
|
| 10 |
"artifact": "scripts/omni/score_model_output_probes.py",
|
|
@@ -101,11 +101,11 @@
|
|
| 101 |
"proxy_scored_task_count": 0,
|
| 102 |
"result_record_count": 20,
|
| 103 |
"scope": "128 selected episodes, held-out test",
|
| 104 |
-
"scored_task_count":
|
| 105 |
-
"scoreless_task_count":
|
| 106 |
"status_counts": {
|
| 107 |
-
"not_evaluated_in_verified_package":
|
| 108 |
-
"scored":
|
| 109 |
}
|
| 110 |
},
|
| 111 |
"raw128_neural_mlp": {
|
|
@@ -140,10 +140,10 @@
|
|
| 140 |
"cosmos3_super_reasoner": 13,
|
| 141 |
"metadata128_neural_mlp": 14,
|
| 142 |
"metadata128_simple": 12,
|
| 143 |
-
"qwen3_omni_v6_lora":
|
| 144 |
},
|
| 145 |
"missing_by_status": {
|
| 146 |
-
"not_evaluated_in_verified_package":
|
| 147 |
"not_supported_by_metadata_only_package": 22,
|
| 148 |
"unsupported_without_required_target": 4
|
| 149 |
},
|
|
@@ -183,15 +183,13 @@
|
|
| 183 |
"11 Temporal Order Verification": [
|
| 184 |
"cosmos3_nano_future_window",
|
| 185 |
"cosmos3_super_reasoner",
|
| 186 |
-
"metadata128_neural_mlp"
|
| 187 |
-
"qwen3_omni_v6_lora"
|
| 188 |
],
|
| 189 |
"12 Multimodal Synchronization Detection": [
|
| 190 |
"cosmos3_nano_future_window",
|
| 191 |
"cosmos3_super_reasoner",
|
| 192 |
"metadata128_neural_mlp",
|
| 193 |
-
"metadata128_simple"
|
| 194 |
-
"qwen3_omni_v6_lora"
|
| 195 |
],
|
| 196 |
"13 Long-Horizon Next-Action Forecasting": [
|
| 197 |
"cosmos3_nano_future_window",
|
|
@@ -241,8 +239,7 @@
|
|
| 241 |
"cosmos3_nano_future_window",
|
| 242 |
"cosmos3_super_reasoner",
|
| 243 |
"metadata128_neural_mlp",
|
| 244 |
-
"metadata128_simple"
|
| 245 |
-
"qwen3_omni_v6_lora"
|
| 246 |
]
|
| 247 |
},
|
| 248 |
"missing_records": [
|
|
@@ -519,19 +516,6 @@
|
|
| 519 |
"task_label": "Temporal Order Verification",
|
| 520 |
"task_number": 11
|
| 521 |
},
|
| 522 |
-
{
|
| 523 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 524 |
-
"metric_key": "f1",
|
| 525 |
-
"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",
|
| 526 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 527 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 528 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 529 |
-
"status": "not_evaluated_in_verified_package",
|
| 530 |
-
"status_label": "not evaluated",
|
| 531 |
-
"task_id": "temporal_order",
|
| 532 |
-
"task_label": "Temporal Order Verification",
|
| 533 |
-
"task_number": 11
|
| 534 |
-
},
|
| 535 |
{
|
| 536 |
"method": "Cosmos3-Super Reasoner",
|
| 537 |
"metric_key": "f1",
|
|
@@ -584,19 +568,6 @@
|
|
| 584 |
"task_label": "Multimodal Synchronization Detection",
|
| 585 |
"task_number": 12
|
| 586 |
},
|
| 587 |
-
{
|
| 588 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 589 |
-
"metric_key": "f1",
|
| 590 |
-
"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",
|
| 591 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 592 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 593 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 594 |
-
"status": "not_evaluated_in_verified_package",
|
| 595 |
-
"status_label": "not evaluated",
|
| 596 |
-
"task_id": "misalignment_detection",
|
| 597 |
-
"task_label": "Multimodal Synchronization Detection",
|
| 598 |
-
"task_number": 12
|
| 599 |
-
},
|
| 600 |
{
|
| 601 |
"method": "Cosmos3-Super Reasoner",
|
| 602 |
"metric_key": "f1",
|
|
@@ -1039,19 +1010,6 @@
|
|
| 1039 |
"task_label": "Time-to-Next-Transition Regression",
|
| 1040 |
"task_number": 20
|
| 1041 |
},
|
| 1042 |
-
{
|
| 1043 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 1044 |
-
"metric_key": "mae",
|
| 1045 |
-
"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",
|
| 1046 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 1047 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1048 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 1049 |
-
"status": "not_evaluated_in_verified_package",
|
| 1050 |
-
"status_label": "not evaluated",
|
| 1051 |
-
"task_id": "time_to_transition",
|
| 1052 |
-
"task_label": "Time-to-Next-Transition Regression",
|
| 1053 |
-
"task_number": 20
|
| 1054 |
-
},
|
| 1055 |
{
|
| 1056 |
"method": "Cosmos3-Super Reasoner",
|
| 1057 |
"metric_key": "mae",
|
|
@@ -1125,8 +1083,8 @@
|
|
| 1125 |
"method_count": 9,
|
| 1126 |
"method_task_record_count": 180,
|
| 1127 |
"proxy_scored_method_task_count": 4,
|
| 1128 |
-
"scored_method_task_count":
|
| 1129 |
-
"scoreless_method_task_count":
|
| 1130 |
"task_count": 20
|
| 1131 |
},
|
| 1132 |
"source_matrix": "docs/data/task_method_20_result_matrix.json",
|
|
|
|
| 1 |
{
|
| 2 |
+
"generated_at_utc": "2026-06-17T21:17:51+00:00",
|
| 3 |
"immediate_actions": [
|
| 4 |
{
|
| 5 |
"artifact": "docs/data/task_method_20_gap_audit.json",
|
| 6 |
"id": "gap_audit",
|
| 7 |
+
"purpose": "Keep the 61 scoreless cells visible and reproducible."
|
| 8 |
},
|
| 9 |
{
|
| 10 |
"artifact": "scripts/omni/score_model_output_probes.py",
|
|
|
|
| 101 |
"proxy_scored_task_count": 0,
|
| 102 |
"result_record_count": 20,
|
| 103 |
"scope": "128 selected episodes, held-out test",
|
| 104 |
+
"scored_task_count": 13,
|
| 105 |
+
"scoreless_task_count": 7,
|
| 106 |
"status_counts": {
|
| 107 |
+
"not_evaluated_in_verified_package": 7,
|
| 108 |
+
"scored": 13
|
| 109 |
}
|
| 110 |
},
|
| 111 |
"raw128_neural_mlp": {
|
|
|
|
| 140 |
"cosmos3_super_reasoner": 13,
|
| 141 |
"metadata128_neural_mlp": 14,
|
| 142 |
"metadata128_simple": 12,
|
| 143 |
+
"qwen3_omni_v6_lora": 7
|
| 144 |
},
|
| 145 |
"missing_by_status": {
|
| 146 |
+
"not_evaluated_in_verified_package": 35,
|
| 147 |
"not_supported_by_metadata_only_package": 22,
|
| 148 |
"unsupported_without_required_target": 4
|
| 149 |
},
|
|
|
|
| 183 |
"11 Temporal Order Verification": [
|
| 184 |
"cosmos3_nano_future_window",
|
| 185 |
"cosmos3_super_reasoner",
|
| 186 |
+
"metadata128_neural_mlp"
|
|
|
|
| 187 |
],
|
| 188 |
"12 Multimodal Synchronization Detection": [
|
| 189 |
"cosmos3_nano_future_window",
|
| 190 |
"cosmos3_super_reasoner",
|
| 191 |
"metadata128_neural_mlp",
|
| 192 |
+
"metadata128_simple"
|
|
|
|
| 193 |
],
|
| 194 |
"13 Long-Horizon Next-Action Forecasting": [
|
| 195 |
"cosmos3_nano_future_window",
|
|
|
|
| 239 |
"cosmos3_nano_future_window",
|
| 240 |
"cosmos3_super_reasoner",
|
| 241 |
"metadata128_neural_mlp",
|
| 242 |
+
"metadata128_simple"
|
|
|
|
| 243 |
]
|
| 244 |
},
|
| 245 |
"missing_records": [
|
|
|
|
| 516 |
"task_label": "Temporal Order Verification",
|
| 517 |
"task_number": 11
|
| 518 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 519 |
{
|
| 520 |
"method": "Cosmos3-Super Reasoner",
|
| 521 |
"metric_key": "f1",
|
|
|
|
| 568 |
"task_label": "Multimodal Synchronization Detection",
|
| 569 |
"task_number": 12
|
| 570 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 571 |
{
|
| 572 |
"method": "Cosmos3-Super Reasoner",
|
| 573 |
"metric_key": "f1",
|
|
|
|
| 1010 |
"task_label": "Time-to-Next-Transition Regression",
|
| 1011 |
"task_number": 20
|
| 1012 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1013 |
{
|
| 1014 |
"method": "Cosmos3-Super Reasoner",
|
| 1015 |
"metric_key": "mae",
|
|
|
|
| 1083 |
"method_count": 9,
|
| 1084 |
"method_task_record_count": 180,
|
| 1085 |
"proxy_scored_method_task_count": 4,
|
| 1086 |
+
"scored_method_task_count": 119,
|
| 1087 |
+
"scoreless_method_task_count": 61,
|
| 1088 |
"task_count": 20
|
| 1089 |
},
|
| 1090 |
"source_matrix": "docs/data/task_method_20_result_matrix.json",
|
data/task_method_20_result_matrix.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
"title": "Task Method 20-Result Matrix",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
-
"scored_method_task_count":
|
| 9 |
"series": [
|
| 10 |
{
|
| 11 |
"id": "minimal",
|
|
@@ -161,17 +161,17 @@
|
|
| 161 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 162 |
"plotted_as": "colored point overlay",
|
| 163 |
"result_record_count": 20,
|
| 164 |
-
"scored_task_count":
|
| 165 |
-
"covered_task_count":
|
| 166 |
"proxy_scored_task_count": 0,
|
| 167 |
-
"scoreless_task_count":
|
| 168 |
"unsupported_task_count": 0,
|
| 169 |
-
"not_evaluated_task_count":
|
| 170 |
"status_counts": {
|
| 171 |
-
"not_evaluated_in_verified_package":
|
| 172 |
-
"scored":
|
| 173 |
},
|
| 174 |
-
"coverage_fraction": 0.
|
| 175 |
"result_record_fraction": 1.0
|
| 176 |
},
|
| 177 |
{
|
|
@@ -1958,17 +1958,17 @@
|
|
| 1958 |
"task_label": "Temporal Order Verification",
|
| 1959 |
"series_id": "qwen3_omni_v6_lora",
|
| 1960 |
"method": "Qwen3-Omni v6 LoRA",
|
| 1961 |
-
"status": "
|
| 1962 |
-
"status_label": "
|
| 1963 |
-
"scored":
|
| 1964 |
"proxy_scored": false,
|
| 1965 |
-
"raw":
|
| 1966 |
-
"raw_text": "
|
| 1967 |
-
"normalized_score":
|
| 1968 |
-
"metric_key": "
|
| 1969 |
-
"source":
|
| 1970 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1971 |
-
"reason":
|
| 1972 |
},
|
| 1973 |
{
|
| 1974 |
"task_number": 11,
|
|
@@ -2120,17 +2120,17 @@
|
|
| 2120 |
"task_label": "Multimodal Synchronization Detection",
|
| 2121 |
"series_id": "qwen3_omni_v6_lora",
|
| 2122 |
"method": "Qwen3-Omni v6 LoRA",
|
| 2123 |
-
"status": "
|
| 2124 |
-
"status_label": "
|
| 2125 |
-
"scored":
|
| 2126 |
"proxy_scored": false,
|
| 2127 |
-
"raw":
|
| 2128 |
-
"raw_text": "
|
| 2129 |
-
"normalized_score":
|
| 2130 |
-
"metric_key": "
|
| 2131 |
-
"source":
|
| 2132 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 2133 |
-
"reason":
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
"task_number": 12,
|
|
@@ -3416,17 +3416,17 @@
|
|
| 3416 |
"task_label": "Time-to-Next-Transition Regression",
|
| 3417 |
"series_id": "qwen3_omni_v6_lora",
|
| 3418 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3419 |
-
"status": "
|
| 3420 |
-
"status_label": "
|
| 3421 |
-
"scored":
|
| 3422 |
"proxy_scored": false,
|
| 3423 |
-
"raw":
|
| 3424 |
-
"raw_text": "
|
| 3425 |
-
"normalized_score":
|
| 3426 |
-
"metric_key": "
|
| 3427 |
-
"source":
|
| 3428 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3429 |
-
"reason":
|
| 3430 |
},
|
| 3431 |
{
|
| 3432 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Task Method 20-Result Matrix",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
+
"scored_method_task_count": 119,
|
| 9 |
"series": [
|
| 10 |
{
|
| 11 |
"id": "minimal",
|
|
|
|
| 161 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 162 |
"plotted_as": "colored point overlay",
|
| 163 |
"result_record_count": 20,
|
| 164 |
+
"scored_task_count": 13,
|
| 165 |
+
"covered_task_count": 13,
|
| 166 |
"proxy_scored_task_count": 0,
|
| 167 |
+
"scoreless_task_count": 7,
|
| 168 |
"unsupported_task_count": 0,
|
| 169 |
+
"not_evaluated_task_count": 7,
|
| 170 |
"status_counts": {
|
| 171 |
+
"not_evaluated_in_verified_package": 7,
|
| 172 |
+
"scored": 13
|
| 173 |
},
|
| 174 |
+
"coverage_fraction": 0.65,
|
| 175 |
"result_record_fraction": 1.0
|
| 176 |
},
|
| 177 |
{
|
|
|
|
| 1958 |
"task_label": "Temporal Order Verification",
|
| 1959 |
"series_id": "qwen3_omni_v6_lora",
|
| 1960 |
"method": "Qwen3-Omni v6 LoRA",
|
| 1961 |
+
"status": "scored",
|
| 1962 |
+
"status_label": "scored",
|
| 1963 |
+
"scored": true,
|
| 1964 |
"proxy_scored": false,
|
| 1965 |
+
"raw": 0.40984631701404173,
|
| 1966 |
+
"raw_text": "0.4098",
|
| 1967 |
+
"normalized_score": 0.40984631701404173,
|
| 1968 |
+
"metric_key": "temporal_order_f1",
|
| 1969 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1970 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1971 |
+
"reason": null
|
| 1972 |
},
|
| 1973 |
{
|
| 1974 |
"task_number": 11,
|
|
|
|
| 2120 |
"task_label": "Multimodal Synchronization Detection",
|
| 2121 |
"series_id": "qwen3_omni_v6_lora",
|
| 2122 |
"method": "Qwen3-Omni v6 LoRA",
|
| 2123 |
+
"status": "scored",
|
| 2124 |
+
"status_label": "scored",
|
| 2125 |
+
"scored": true,
|
| 2126 |
"proxy_scored": false,
|
| 2127 |
+
"raw": 0.3344936184319576,
|
| 2128 |
+
"raw_text": "0.3345",
|
| 2129 |
+
"normalized_score": 0.3344936184319576,
|
| 2130 |
+
"metric_key": "misalignment_detection_f1",
|
| 2131 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 2132 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 2133 |
+
"reason": null
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
"task_number": 12,
|
|
|
|
| 3416 |
"task_label": "Time-to-Next-Transition Regression",
|
| 3417 |
"series_id": "qwen3_omni_v6_lora",
|
| 3418 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3419 |
+
"status": "scored",
|
| 3420 |
+
"status_label": "scored",
|
| 3421 |
+
"scored": true,
|
| 3422 |
"proxy_scored": false,
|
| 3423 |
+
"raw": 134.0687422166874,
|
| 3424 |
+
"raw_text": "134.07",
|
| 3425 |
+
"normalized_score": 0.07859666766782253,
|
| 3426 |
+
"metric_key": "time_to_transition_mae",
|
| 3427 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 3428 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3429 |
+
"reason": null
|
| 3430 |
},
|
| 3431 |
{
|
| 3432 |
"task_number": 20,
|
data/task_surface_integrity.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-
|
| 4 |
"summary": {
|
| 5 |
"task_count": 12,
|
| 6 |
"expected_task_count": 12,
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:26+00:00",
|
| 4 |
"summary": {
|
| 5 |
"task_count": 12,
|
| 6 |
"expected_task_count": 12,
|
data/unified_task_model_radar.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
"title": "Unified 20-Task Model Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
-
"scored_method_task_count":
|
| 9 |
"normalization_policy": {
|
| 10 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 11 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
@@ -170,17 +170,17 @@
|
|
| 170 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 171 |
"plotted_as": "colored point overlay",
|
| 172 |
"result_record_count": 20,
|
| 173 |
-
"scored_task_count":
|
| 174 |
-
"covered_task_count":
|
| 175 |
"proxy_scored_task_count": 0,
|
| 176 |
-
"scoreless_task_count":
|
| 177 |
"unsupported_task_count": 0,
|
| 178 |
-
"not_evaluated_task_count":
|
| 179 |
"status_counts": {
|
| 180 |
-
"not_evaluated_in_verified_package":
|
| 181 |
-
"scored":
|
| 182 |
},
|
| 183 |
-
"coverage_fraction": 0.
|
| 184 |
"result_record_fraction": 1.0
|
| 185 |
},
|
| 186 |
{
|
|
@@ -1375,6 +1375,17 @@
|
|
| 1375 |
"raw_text": "0.8520",
|
| 1376 |
"status_label": "scored"
|
| 1377 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1378 |
"metadata128_simple": {
|
| 1379 |
"raw": 0.4198864140782312,
|
| 1380 |
"metric_key": "f1",
|
|
@@ -1419,17 +1430,6 @@
|
|
| 1419 |
"raw_text": "n/a",
|
| 1420 |
"status_label": "not supported"
|
| 1421 |
},
|
| 1422 |
-
"qwen3_omni_v6_lora": {
|
| 1423 |
-
"raw": null,
|
| 1424 |
-
"metric_key": "f1",
|
| 1425 |
-
"source": null,
|
| 1426 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1427 |
-
"status": "not_evaluated_in_verified_package",
|
| 1428 |
-
"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",
|
| 1429 |
-
"normalized_score": null,
|
| 1430 |
-
"raw_text": "n/a",
|
| 1431 |
-
"status_label": "not evaluated"
|
| 1432 |
-
},
|
| 1433 |
"cosmos3_super_reasoner": {
|
| 1434 |
"raw": null,
|
| 1435 |
"metric_key": "f1",
|
|
@@ -1486,6 +1486,17 @@
|
|
| 1486 |
"raw_text": "0.7153",
|
| 1487 |
"status_label": "scored"
|
| 1488 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1489 |
"metadata128_simple": {
|
| 1490 |
"raw": null,
|
| 1491 |
"metric_key": "f1",
|
|
@@ -1530,17 +1541,6 @@
|
|
| 1530 |
"raw_text": "n/a",
|
| 1531 |
"status_label": "not supported"
|
| 1532 |
},
|
| 1533 |
-
"qwen3_omni_v6_lora": {
|
| 1534 |
-
"raw": null,
|
| 1535 |
-
"metric_key": "f1",
|
| 1536 |
-
"source": null,
|
| 1537 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1538 |
-
"status": "not_evaluated_in_verified_package",
|
| 1539 |
-
"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",
|
| 1540 |
-
"normalized_score": null,
|
| 1541 |
-
"raw_text": "n/a",
|
| 1542 |
-
"status_label": "not evaluated"
|
| 1543 |
-
},
|
| 1544 |
"cosmos3_super_reasoner": {
|
| 1545 |
"raw": null,
|
| 1546 |
"metric_key": "f1",
|
|
@@ -2374,6 +2374,17 @@
|
|
| 2374 |
"raw_text": "10.55",
|
| 2375 |
"status_label": "scored"
|
| 2376 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2377 |
"raw128_simple": {
|
| 2378 |
"raw": 52.32759475708008,
|
| 2379 |
"metric_key": "mae",
|
|
@@ -2418,17 +2429,6 @@
|
|
| 2418 |
"raw_text": "n/a",
|
| 2419 |
"status_label": "not supported"
|
| 2420 |
},
|
| 2421 |
-
"qwen3_omni_v6_lora": {
|
| 2422 |
-
"raw": null,
|
| 2423 |
-
"metric_key": "mae",
|
| 2424 |
-
"source": null,
|
| 2425 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 2426 |
-
"status": "not_evaluated_in_verified_package",
|
| 2427 |
-
"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",
|
| 2428 |
-
"normalized_score": null,
|
| 2429 |
-
"raw_text": "n/a",
|
| 2430 |
-
"status_label": "not evaluated"
|
| 2431 |
-
},
|
| 2432 |
"cosmos3_super_reasoner": {
|
| 2433 |
"raw": null,
|
| 2434 |
"metric_key": "mae",
|
|
@@ -2492,7 +2492,7 @@
|
|
| 2492 |
"title": "Qwen3-Omni v6 LoRA",
|
| 2493 |
"status": "verified",
|
| 2494 |
"task_aligned_axes": "Qwen3",
|
| 2495 |
-
"coverage": "20 records /
|
| 2496 |
"headline": "JSON validity 0.9990; action macro-F1 0.0029",
|
| 2497 |
"source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_multiscale_cap96_v6_rank64_lr5e5_full8gpu_lora_eval_test_full/eval/metrics.json"
|
| 2498 |
},
|
|
@@ -4256,17 +4256,17 @@
|
|
| 4256 |
"task_label": "Temporal Order Verification",
|
| 4257 |
"series_id": "qwen3_omni_v6_lora",
|
| 4258 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4259 |
-
"status": "
|
| 4260 |
-
"status_label": "
|
| 4261 |
-
"scored":
|
| 4262 |
"proxy_scored": false,
|
| 4263 |
-
"raw":
|
| 4264 |
-
"raw_text": "
|
| 4265 |
-
"normalized_score":
|
| 4266 |
-
"metric_key": "
|
| 4267 |
-
"source":
|
| 4268 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4269 |
-
"reason":
|
| 4270 |
},
|
| 4271 |
{
|
| 4272 |
"task_number": 11,
|
|
@@ -4418,17 +4418,17 @@
|
|
| 4418 |
"task_label": "Multimodal Synchronization Detection",
|
| 4419 |
"series_id": "qwen3_omni_v6_lora",
|
| 4420 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4421 |
-
"status": "
|
| 4422 |
-
"status_label": "
|
| 4423 |
-
"scored":
|
| 4424 |
"proxy_scored": false,
|
| 4425 |
-
"raw":
|
| 4426 |
-
"raw_text": "
|
| 4427 |
-
"normalized_score":
|
| 4428 |
-
"metric_key": "
|
| 4429 |
-
"source":
|
| 4430 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4431 |
-
"reason":
|
| 4432 |
},
|
| 4433 |
{
|
| 4434 |
"task_number": 12,
|
|
@@ -5714,17 +5714,17 @@
|
|
| 5714 |
"task_label": "Time-to-Next-Transition Regression",
|
| 5715 |
"series_id": "qwen3_omni_v6_lora",
|
| 5716 |
"method": "Qwen3-Omni v6 LoRA",
|
| 5717 |
-
"status": "
|
| 5718 |
-
"status_label": "
|
| 5719 |
-
"scored":
|
| 5720 |
"proxy_scored": false,
|
| 5721 |
-
"raw":
|
| 5722 |
-
"raw_text": "
|
| 5723 |
-
"normalized_score":
|
| 5724 |
-
"metric_key": "
|
| 5725 |
-
"source":
|
| 5726 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 5727 |
-
"reason":
|
| 5728 |
},
|
| 5729 |
{
|
| 5730 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Unified 20-Task Model Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
+
"scored_method_task_count": 119,
|
| 9 |
"normalization_policy": {
|
| 10 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 11 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
|
|
| 170 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 171 |
"plotted_as": "colored point overlay",
|
| 172 |
"result_record_count": 20,
|
| 173 |
+
"scored_task_count": 13,
|
| 174 |
+
"covered_task_count": 13,
|
| 175 |
"proxy_scored_task_count": 0,
|
| 176 |
+
"scoreless_task_count": 7,
|
| 177 |
"unsupported_task_count": 0,
|
| 178 |
+
"not_evaluated_task_count": 7,
|
| 179 |
"status_counts": {
|
| 180 |
+
"not_evaluated_in_verified_package": 7,
|
| 181 |
+
"scored": 13
|
| 182 |
},
|
| 183 |
+
"coverage_fraction": 0.65,
|
| 184 |
"result_record_fraction": 1.0
|
| 185 |
},
|
| 186 |
{
|
|
|
|
| 1375 |
"raw_text": "0.8520",
|
| 1376 |
"status_label": "scored"
|
| 1377 |
},
|
| 1378 |
+
"qwen3_omni_v6_lora": {
|
| 1379 |
+
"raw": 0.40984631701404173,
|
| 1380 |
+
"metric_key": "temporal_order_f1",
|
| 1381 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1382 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 1383 |
+
"status": "scored",
|
| 1384 |
+
"reason": null,
|
| 1385 |
+
"normalized_score": 0.40984631701404173,
|
| 1386 |
+
"raw_text": "0.4098",
|
| 1387 |
+
"status_label": "scored"
|
| 1388 |
+
},
|
| 1389 |
"metadata128_simple": {
|
| 1390 |
"raw": 0.4198864140782312,
|
| 1391 |
"metric_key": "f1",
|
|
|
|
| 1430 |
"raw_text": "n/a",
|
| 1431 |
"status_label": "not supported"
|
| 1432 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1433 |
"cosmos3_super_reasoner": {
|
| 1434 |
"raw": null,
|
| 1435 |
"metric_key": "f1",
|
|
|
|
| 1486 |
"raw_text": "0.7153",
|
| 1487 |
"status_label": "scored"
|
| 1488 |
},
|
| 1489 |
+
"qwen3_omni_v6_lora": {
|
| 1490 |
+
"raw": 0.3344936184319576,
|
| 1491 |
+
"metric_key": "misalignment_detection_f1",
|
| 1492 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 1493 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 1494 |
+
"status": "scored",
|
| 1495 |
+
"reason": null,
|
| 1496 |
+
"normalized_score": 0.3344936184319576,
|
| 1497 |
+
"raw_text": "0.3345",
|
| 1498 |
+
"status_label": "scored"
|
| 1499 |
+
},
|
| 1500 |
"metadata128_simple": {
|
| 1501 |
"raw": null,
|
| 1502 |
"metric_key": "f1",
|
|
|
|
| 1541 |
"raw_text": "n/a",
|
| 1542 |
"status_label": "not supported"
|
| 1543 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1544 |
"cosmos3_super_reasoner": {
|
| 1545 |
"raw": null,
|
| 1546 |
"metric_key": "f1",
|
|
|
|
| 2374 |
"raw_text": "10.55",
|
| 2375 |
"status_label": "scored"
|
| 2376 |
},
|
| 2377 |
+
"qwen3_omni_v6_lora": {
|
| 2378 |
+
"raw": 134.0687422166874,
|
| 2379 |
+
"metric_key": "time_to_transition_mae",
|
| 2380 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 2381 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 2382 |
+
"status": "scored",
|
| 2383 |
+
"reason": null,
|
| 2384 |
+
"normalized_score": 0.07859666766782253,
|
| 2385 |
+
"raw_text": "134.07",
|
| 2386 |
+
"status_label": "scored"
|
| 2387 |
+
},
|
| 2388 |
"raw128_simple": {
|
| 2389 |
"raw": 52.32759475708008,
|
| 2390 |
"metric_key": "mae",
|
|
|
|
| 2429 |
"raw_text": "n/a",
|
| 2430 |
"status_label": "not supported"
|
| 2431 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2432 |
"cosmos3_super_reasoner": {
|
| 2433 |
"raw": null,
|
| 2434 |
"metric_key": "mae",
|
|
|
|
| 2492 |
"title": "Qwen3-Omni v6 LoRA",
|
| 2493 |
"status": "verified",
|
| 2494 |
"task_aligned_axes": "Qwen3",
|
| 2495 |
+
"coverage": "20 records / 13 scored task-aligned axes",
|
| 2496 |
"headline": "JSON validity 0.9990; action macro-F1 0.0029",
|
| 2497 |
"source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_multiscale_cap96_v6_rank64_lr5e5_full8gpu_lora_eval_test_full/eval/metrics.json"
|
| 2498 |
},
|
|
|
|
| 4256 |
"task_label": "Temporal Order Verification",
|
| 4257 |
"series_id": "qwen3_omni_v6_lora",
|
| 4258 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4259 |
+
"status": "scored",
|
| 4260 |
+
"status_label": "scored",
|
| 4261 |
+
"scored": true,
|
| 4262 |
"proxy_scored": false,
|
| 4263 |
+
"raw": 0.40984631701404173,
|
| 4264 |
+
"raw_text": "0.4098",
|
| 4265 |
+
"normalized_score": 0.40984631701404173,
|
| 4266 |
+
"metric_key": "temporal_order_f1",
|
| 4267 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 4268 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4269 |
+
"reason": null
|
| 4270 |
},
|
| 4271 |
{
|
| 4272 |
"task_number": 11,
|
|
|
|
| 4418 |
"task_label": "Multimodal Synchronization Detection",
|
| 4419 |
"series_id": "qwen3_omni_v6_lora",
|
| 4420 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4421 |
+
"status": "scored",
|
| 4422 |
+
"status_label": "scored",
|
| 4423 |
+
"scored": true,
|
| 4424 |
"proxy_scored": false,
|
| 4425 |
+
"raw": 0.3344936184319576,
|
| 4426 |
+
"raw_text": "0.3345",
|
| 4427 |
+
"normalized_score": 0.3344936184319576,
|
| 4428 |
+
"metric_key": "misalignment_detection_f1",
|
| 4429 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 4430 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4431 |
+
"reason": null
|
| 4432 |
},
|
| 4433 |
{
|
| 4434 |
"task_number": 12,
|
|
|
|
| 5714 |
"task_label": "Time-to-Next-Transition Regression",
|
| 5715 |
"series_id": "qwen3_omni_v6_lora",
|
| 5716 |
"method": "Qwen3-Omni v6 LoRA",
|
| 5717 |
+
"status": "scored",
|
| 5718 |
+
"status_label": "scored",
|
| 5719 |
+
"scored": true,
|
| 5720 |
"proxy_scored": false,
|
| 5721 |
+
"raw": 134.0687422166874,
|
| 5722 |
+
"raw_text": "134.07",
|
| 5723 |
+
"normalized_score": 0.07859666766782253,
|
| 5724 |
+
"metric_key": "time_to_transition_mae",
|
| 5725 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 5726 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 5727 |
+
"reason": null
|
| 5728 |
},
|
| 5729 |
{
|
| 5730 |
"task_number": 20,
|
data/website_integrity.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-17T21:
|
| 4 |
"docs_root": "docs",
|
| 5 |
"site_base": "/ropedia-xperience-10m-task-suite/",
|
| 6 |
"summary": {
|
|
@@ -316,7 +316,7 @@
|
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"path": "data/episode128_task_model_radar.json",
|
| 319 |
-
"bytes":
|
| 320 |
"top_level_type": "dict"
|
| 321 |
},
|
| 322 |
{
|
|
@@ -486,12 +486,12 @@
|
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"path": "data/task_method_20_gap_audit.json",
|
| 489 |
-
"bytes":
|
| 490 |
"top_level_type": "dict"
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"path": "data/task_method_20_result_matrix.json",
|
| 494 |
-
"bytes":
|
| 495 |
"top_level_type": "dict"
|
| 496 |
},
|
| 497 |
{
|
|
@@ -526,7 +526,7 @@
|
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"path": "data/unified_task_model_radar.json",
|
| 529 |
-
"bytes":
|
| 530 |
"top_level_type": "dict"
|
| 531 |
},
|
| 532 |
{
|
|
@@ -566,7 +566,7 @@
|
|
| 566 |
{
|
| 567 |
"path": "assets/charts/episode128_task_model_radar.svg",
|
| 568 |
"exists": true,
|
| 569 |
-
"bytes":
|
| 570 |
"format": "SVG",
|
| 571 |
"has_viewbox": true
|
| 572 |
},
|
|
@@ -636,7 +636,7 @@
|
|
| 636 |
{
|
| 637 |
"path": "assets/charts/unified_task_model_radar.svg",
|
| 638 |
"exists": true,
|
| 639 |
-
"bytes":
|
| 640 |
"format": "SVG",
|
| 641 |
"has_viewbox": true
|
| 642 |
},
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:27+00:00",
|
| 4 |
"docs_root": "docs",
|
| 5 |
"site_base": "/ropedia-xperience-10m-task-suite/",
|
| 6 |
"summary": {
|
|
|
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"path": "data/episode128_task_model_radar.json",
|
| 319 |
+
"bytes": 187309,
|
| 320 |
"top_level_type": "dict"
|
| 321 |
},
|
| 322 |
{
|
|
|
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"path": "data/task_method_20_gap_audit.json",
|
| 489 |
+
"bytes": 53574,
|
| 490 |
"top_level_type": "dict"
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"path": "data/task_method_20_result_matrix.json",
|
| 494 |
+
"bytes": 129707,
|
| 495 |
"top_level_type": "dict"
|
| 496 |
},
|
| 497 |
{
|
|
|
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"path": "data/unified_task_model_radar.json",
|
| 529 |
+
"bytes": 231161,
|
| 530 |
"top_level_type": "dict"
|
| 531 |
},
|
| 532 |
{
|
|
|
|
| 566 |
{
|
| 567 |
"path": "assets/charts/episode128_task_model_radar.svg",
|
| 568 |
"exists": true,
|
| 569 |
+
"bytes": 44378,
|
| 570 |
"format": "SVG",
|
| 571 |
"has_viewbox": true
|
| 572 |
},
|
|
|
|
| 636 |
{
|
| 637 |
"path": "assets/charts/unified_task_model_radar.svg",
|
| 638 |
"exists": true,
|
| 639 |
+
"bytes": 50394,
|
| 640 |
"format": "SVG",
|
| 641 |
"has_viewbox": true
|
| 642 |
},
|
docs/data/episode128_task_model_radar.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"title": "128-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"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.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 7,
|
| 8 |
"method_task_record_count": 140,
|
| 9 |
-
"scored_method_task_count":
|
| 10 |
"normalization_policy": {
|
| 11 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 12 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
@@ -127,17 +127,17 @@
|
|
| 127 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 128 |
"plotted_as": "colored point overlay",
|
| 129 |
"result_record_count": 20,
|
| 130 |
-
"scored_task_count":
|
| 131 |
-
"covered_task_count":
|
| 132 |
"proxy_scored_task_count": 0,
|
| 133 |
-
"scoreless_task_count":
|
| 134 |
"unsupported_task_count": 0,
|
| 135 |
-
"not_evaluated_task_count":
|
| 136 |
"status_counts": {
|
| 137 |
-
"not_evaluated_in_verified_package":
|
| 138 |
-
"scored":
|
| 139 |
},
|
| 140 |
-
"coverage_fraction": 0.
|
| 141 |
"result_record_fraction": 1.0
|
| 142 |
},
|
| 143 |
{
|
|
@@ -1157,15 +1157,15 @@
|
|
| 1157 |
"status_label": "scored"
|
| 1158 |
},
|
| 1159 |
"qwen3_omni_v6_lora": {
|
| 1160 |
-
"raw":
|
| 1161 |
-
"metric_key": "
|
| 1162 |
-
"source":
|
| 1163 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1164 |
-
"status": "
|
| 1165 |
-
"reason":
|
| 1166 |
-
"normalized_score":
|
| 1167 |
-
"raw_text": "
|
| 1168 |
-
"status_label": "
|
| 1169 |
},
|
| 1170 |
"cosmos3_super_reasoner": {
|
| 1171 |
"raw": null,
|
|
@@ -1248,15 +1248,15 @@
|
|
| 1248 |
"status_label": "scored"
|
| 1249 |
},
|
| 1250 |
"qwen3_omni_v6_lora": {
|
| 1251 |
-
"raw":
|
| 1252 |
-
"metric_key": "
|
| 1253 |
-
"source":
|
| 1254 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1255 |
-
"status": "
|
| 1256 |
-
"reason":
|
| 1257 |
-
"normalized_score":
|
| 1258 |
-
"raw_text": "
|
| 1259 |
-
"status_label": "
|
| 1260 |
},
|
| 1261 |
"cosmos3_super_reasoner": {
|
| 1262 |
"raw": null,
|
|
@@ -1976,15 +1976,15 @@
|
|
| 1976 |
"status_label": "scored"
|
| 1977 |
},
|
| 1978 |
"qwen3_omni_v6_lora": {
|
| 1979 |
-
"raw":
|
| 1980 |
-
"metric_key": "
|
| 1981 |
-
"source":
|
| 1982 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1983 |
-
"status": "
|
| 1984 |
-
"reason":
|
| 1985 |
-
"normalized_score":
|
| 1986 |
-
"raw_text": "
|
| 1987 |
-
"status_label": "
|
| 1988 |
},
|
| 1989 |
"cosmos3_super_reasoner": {
|
| 1990 |
"raw": null,
|
|
@@ -3350,17 +3350,17 @@
|
|
| 3350 |
"task_label": "Temporal Order Verification",
|
| 3351 |
"series_id": "qwen3_omni_v6_lora",
|
| 3352 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3353 |
-
"status": "
|
| 3354 |
-
"status_label": "
|
| 3355 |
-
"scored":
|
| 3356 |
"proxy_scored": false,
|
| 3357 |
-
"raw":
|
| 3358 |
-
"raw_text": "
|
| 3359 |
-
"normalized_score":
|
| 3360 |
-
"metric_key": "
|
| 3361 |
-
"source":
|
| 3362 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3363 |
-
"reason":
|
| 3364 |
},
|
| 3365 |
{
|
| 3366 |
"task_number": 11,
|
|
@@ -3476,17 +3476,17 @@
|
|
| 3476 |
"task_label": "Multimodal Synchronization Detection",
|
| 3477 |
"series_id": "qwen3_omni_v6_lora",
|
| 3478 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3479 |
-
"status": "
|
| 3480 |
-
"status_label": "
|
| 3481 |
-
"scored":
|
| 3482 |
"proxy_scored": false,
|
| 3483 |
-
"raw":
|
| 3484 |
-
"raw_text": "
|
| 3485 |
-
"normalized_score":
|
| 3486 |
-
"metric_key": "
|
| 3487 |
-
"source":
|
| 3488 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3489 |
-
"reason":
|
| 3490 |
},
|
| 3491 |
{
|
| 3492 |
"task_number": 12,
|
|
@@ -4484,17 +4484,17 @@
|
|
| 4484 |
"task_label": "Time-to-Next-Transition Regression",
|
| 4485 |
"series_id": "qwen3_omni_v6_lora",
|
| 4486 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4487 |
-
"status": "
|
| 4488 |
-
"status_label": "
|
| 4489 |
-
"scored":
|
| 4490 |
"proxy_scored": false,
|
| 4491 |
-
"raw":
|
| 4492 |
-
"raw_text": "
|
| 4493 |
-
"normalized_score":
|
| 4494 |
-
"metric_key": "
|
| 4495 |
-
"source":
|
| 4496 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4497 |
-
"reason":
|
| 4498 |
},
|
| 4499 |
{
|
| 4500 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "128-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"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.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 7,
|
| 8 |
"method_task_record_count": 140,
|
| 9 |
+
"scored_method_task_count": 79,
|
| 10 |
"normalization_policy": {
|
| 11 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 12 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
|
|
| 127 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 128 |
"plotted_as": "colored point overlay",
|
| 129 |
"result_record_count": 20,
|
| 130 |
+
"scored_task_count": 13,
|
| 131 |
+
"covered_task_count": 13,
|
| 132 |
"proxy_scored_task_count": 0,
|
| 133 |
+
"scoreless_task_count": 7,
|
| 134 |
"unsupported_task_count": 0,
|
| 135 |
+
"not_evaluated_task_count": 7,
|
| 136 |
"status_counts": {
|
| 137 |
+
"not_evaluated_in_verified_package": 7,
|
| 138 |
+
"scored": 13
|
| 139 |
},
|
| 140 |
+
"coverage_fraction": 0.65,
|
| 141 |
"result_record_fraction": 1.0
|
| 142 |
},
|
| 143 |
{
|
|
|
|
| 1157 |
"status_label": "scored"
|
| 1158 |
},
|
| 1159 |
"qwen3_omni_v6_lora": {
|
| 1160 |
+
"raw": 0.40984631701404173,
|
| 1161 |
+
"metric_key": "temporal_order_f1",
|
| 1162 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1163 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1164 |
+
"status": "scored",
|
| 1165 |
+
"reason": null,
|
| 1166 |
+
"normalized_score": 0.40984631701404173,
|
| 1167 |
+
"raw_text": "0.4098",
|
| 1168 |
+
"status_label": "scored"
|
| 1169 |
},
|
| 1170 |
"cosmos3_super_reasoner": {
|
| 1171 |
"raw": null,
|
|
|
|
| 1248 |
"status_label": "scored"
|
| 1249 |
},
|
| 1250 |
"qwen3_omni_v6_lora": {
|
| 1251 |
+
"raw": 0.3344936184319576,
|
| 1252 |
+
"metric_key": "misalignment_detection_f1",
|
| 1253 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 1254 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1255 |
+
"status": "scored",
|
| 1256 |
+
"reason": null,
|
| 1257 |
+
"normalized_score": 0.3344936184319576,
|
| 1258 |
+
"raw_text": "0.3345",
|
| 1259 |
+
"status_label": "scored"
|
| 1260 |
},
|
| 1261 |
"cosmos3_super_reasoner": {
|
| 1262 |
"raw": null,
|
|
|
|
| 1976 |
"status_label": "scored"
|
| 1977 |
},
|
| 1978 |
"qwen3_omni_v6_lora": {
|
| 1979 |
+
"raw": 134.0687422166874,
|
| 1980 |
+
"metric_key": "time_to_transition_mae",
|
| 1981 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 1982 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1983 |
+
"status": "scored",
|
| 1984 |
+
"reason": null,
|
| 1985 |
+
"normalized_score": 0.07859666766782253,
|
| 1986 |
+
"raw_text": "134.07",
|
| 1987 |
+
"status_label": "scored"
|
| 1988 |
},
|
| 1989 |
"cosmos3_super_reasoner": {
|
| 1990 |
"raw": null,
|
|
|
|
| 3350 |
"task_label": "Temporal Order Verification",
|
| 3351 |
"series_id": "qwen3_omni_v6_lora",
|
| 3352 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3353 |
+
"status": "scored",
|
| 3354 |
+
"status_label": "scored",
|
| 3355 |
+
"scored": true,
|
| 3356 |
"proxy_scored": false,
|
| 3357 |
+
"raw": 0.40984631701404173,
|
| 3358 |
+
"raw_text": "0.4098",
|
| 3359 |
+
"normalized_score": 0.40984631701404173,
|
| 3360 |
+
"metric_key": "temporal_order_f1",
|
| 3361 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 3362 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3363 |
+
"reason": null
|
| 3364 |
},
|
| 3365 |
{
|
| 3366 |
"task_number": 11,
|
|
|
|
| 3476 |
"task_label": "Multimodal Synchronization Detection",
|
| 3477 |
"series_id": "qwen3_omni_v6_lora",
|
| 3478 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3479 |
+
"status": "scored",
|
| 3480 |
+
"status_label": "scored",
|
| 3481 |
+
"scored": true,
|
| 3482 |
"proxy_scored": false,
|
| 3483 |
+
"raw": 0.3344936184319576,
|
| 3484 |
+
"raw_text": "0.3345",
|
| 3485 |
+
"normalized_score": 0.3344936184319576,
|
| 3486 |
+
"metric_key": "misalignment_detection_f1",
|
| 3487 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 3488 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3489 |
+
"reason": null
|
| 3490 |
},
|
| 3491 |
{
|
| 3492 |
"task_number": 12,
|
|
|
|
| 4484 |
"task_label": "Time-to-Next-Transition Regression",
|
| 4485 |
"series_id": "qwen3_omni_v6_lora",
|
| 4486 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4487 |
+
"status": "scored",
|
| 4488 |
+
"status_label": "scored",
|
| 4489 |
+
"scored": true,
|
| 4490 |
"proxy_scored": false,
|
| 4491 |
+
"raw": 134.0687422166874,
|
| 4492 |
+
"raw_text": "134.07",
|
| 4493 |
+
"normalized_score": 0.07859666766782253,
|
| 4494 |
+
"metric_key": "time_to_transition_mae",
|
| 4495 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 4496 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4497 |
+
"reason": null
|
| 4498 |
},
|
| 4499 |
{
|
| 4500 |
"task_number": 20,
|
docs/data/publication_audit.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-17T21:
|
| 4 |
"checks": [
|
| 5 |
{
|
| 6 |
"name": "required_publication_assets_present",
|
|
@@ -206,8 +206,8 @@
|
|
| 206 |
"github_repo": {
|
| 207 |
"root": "repo",
|
| 208 |
"exists": true,
|
| 209 |
-
"file_count":
|
| 210 |
-
"text_file_count":
|
| 211 |
"largest_file": {
|
| 212 |
"path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
|
| 213 |
"bytes": 55702978
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:35+00:00",
|
| 4 |
"checks": [
|
| 5 |
{
|
| 6 |
"name": "required_publication_assets_present",
|
|
|
|
| 206 |
"github_repo": {
|
| 207 |
"root": "repo",
|
| 208 |
"exists": true,
|
| 209 |
+
"file_count": 1250,
|
| 210 |
+
"text_file_count": 1052,
|
| 211 |
"largest_file": {
|
| 212 |
"path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
|
| 213 |
"bytes": 55702978
|
docs/data/single_episode_task_model_radar.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"title": "Single-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"description": "Minimal and Neural MLP baselines on the one public sample episode, both scored on all 20 task contracts.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 2,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Single-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"description": "Minimal and Neural MLP baselines on the one public sample episode, both scored on all 20 task contracts.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 2,
|
docs/data/task_method_20_gap_audit.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"generated_at_utc": "2026-06-
|
| 3 |
"immediate_actions": [
|
| 4 |
{
|
| 5 |
"artifact": "docs/data/task_method_20_gap_audit.json",
|
| 6 |
"id": "gap_audit",
|
| 7 |
-
"purpose": "Keep the
|
| 8 |
},
|
| 9 |
{
|
| 10 |
"artifact": "scripts/omni/score_model_output_probes.py",
|
|
@@ -101,11 +101,11 @@
|
|
| 101 |
"proxy_scored_task_count": 0,
|
| 102 |
"result_record_count": 20,
|
| 103 |
"scope": "128 selected episodes, held-out test",
|
| 104 |
-
"scored_task_count":
|
| 105 |
-
"scoreless_task_count":
|
| 106 |
"status_counts": {
|
| 107 |
-
"not_evaluated_in_verified_package":
|
| 108 |
-
"scored":
|
| 109 |
}
|
| 110 |
},
|
| 111 |
"raw128_neural_mlp": {
|
|
@@ -140,10 +140,10 @@
|
|
| 140 |
"cosmos3_super_reasoner": 13,
|
| 141 |
"metadata128_neural_mlp": 14,
|
| 142 |
"metadata128_simple": 12,
|
| 143 |
-
"qwen3_omni_v6_lora":
|
| 144 |
},
|
| 145 |
"missing_by_status": {
|
| 146 |
-
"not_evaluated_in_verified_package":
|
| 147 |
"not_supported_by_metadata_only_package": 22,
|
| 148 |
"unsupported_without_required_target": 4
|
| 149 |
},
|
|
@@ -183,15 +183,13 @@
|
|
| 183 |
"11 Temporal Order Verification": [
|
| 184 |
"cosmos3_nano_future_window",
|
| 185 |
"cosmos3_super_reasoner",
|
| 186 |
-
"metadata128_neural_mlp"
|
| 187 |
-
"qwen3_omni_v6_lora"
|
| 188 |
],
|
| 189 |
"12 Multimodal Synchronization Detection": [
|
| 190 |
"cosmos3_nano_future_window",
|
| 191 |
"cosmos3_super_reasoner",
|
| 192 |
"metadata128_neural_mlp",
|
| 193 |
-
"metadata128_simple"
|
| 194 |
-
"qwen3_omni_v6_lora"
|
| 195 |
],
|
| 196 |
"13 Long-Horizon Next-Action Forecasting": [
|
| 197 |
"cosmos3_nano_future_window",
|
|
@@ -241,8 +239,7 @@
|
|
| 241 |
"cosmos3_nano_future_window",
|
| 242 |
"cosmos3_super_reasoner",
|
| 243 |
"metadata128_neural_mlp",
|
| 244 |
-
"metadata128_simple"
|
| 245 |
-
"qwen3_omni_v6_lora"
|
| 246 |
]
|
| 247 |
},
|
| 248 |
"missing_records": [
|
|
@@ -519,19 +516,6 @@
|
|
| 519 |
"task_label": "Temporal Order Verification",
|
| 520 |
"task_number": 11
|
| 521 |
},
|
| 522 |
-
{
|
| 523 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 524 |
-
"metric_key": "f1",
|
| 525 |
-
"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",
|
| 526 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 527 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 528 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 529 |
-
"status": "not_evaluated_in_verified_package",
|
| 530 |
-
"status_label": "not evaluated",
|
| 531 |
-
"task_id": "temporal_order",
|
| 532 |
-
"task_label": "Temporal Order Verification",
|
| 533 |
-
"task_number": 11
|
| 534 |
-
},
|
| 535 |
{
|
| 536 |
"method": "Cosmos3-Super Reasoner",
|
| 537 |
"metric_key": "f1",
|
|
@@ -584,19 +568,6 @@
|
|
| 584 |
"task_label": "Multimodal Synchronization Detection",
|
| 585 |
"task_number": 12
|
| 586 |
},
|
| 587 |
-
{
|
| 588 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 589 |
-
"metric_key": "f1",
|
| 590 |
-
"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",
|
| 591 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 592 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 593 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 594 |
-
"status": "not_evaluated_in_verified_package",
|
| 595 |
-
"status_label": "not evaluated",
|
| 596 |
-
"task_id": "misalignment_detection",
|
| 597 |
-
"task_label": "Multimodal Synchronization Detection",
|
| 598 |
-
"task_number": 12
|
| 599 |
-
},
|
| 600 |
{
|
| 601 |
"method": "Cosmos3-Super Reasoner",
|
| 602 |
"metric_key": "f1",
|
|
@@ -1039,19 +1010,6 @@
|
|
| 1039 |
"task_label": "Time-to-Next-Transition Regression",
|
| 1040 |
"task_number": 20
|
| 1041 |
},
|
| 1042 |
-
{
|
| 1043 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 1044 |
-
"metric_key": "mae",
|
| 1045 |
-
"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",
|
| 1046 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 1047 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1048 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 1049 |
-
"status": "not_evaluated_in_verified_package",
|
| 1050 |
-
"status_label": "not evaluated",
|
| 1051 |
-
"task_id": "time_to_transition",
|
| 1052 |
-
"task_label": "Time-to-Next-Transition Regression",
|
| 1053 |
-
"task_number": 20
|
| 1054 |
-
},
|
| 1055 |
{
|
| 1056 |
"method": "Cosmos3-Super Reasoner",
|
| 1057 |
"metric_key": "mae",
|
|
@@ -1125,8 +1083,8 @@
|
|
| 1125 |
"method_count": 9,
|
| 1126 |
"method_task_record_count": 180,
|
| 1127 |
"proxy_scored_method_task_count": 4,
|
| 1128 |
-
"scored_method_task_count":
|
| 1129 |
-
"scoreless_method_task_count":
|
| 1130 |
"task_count": 20
|
| 1131 |
},
|
| 1132 |
"source_matrix": "docs/data/task_method_20_result_matrix.json",
|
|
|
|
| 1 |
{
|
| 2 |
+
"generated_at_utc": "2026-06-17T21:17:51+00:00",
|
| 3 |
"immediate_actions": [
|
| 4 |
{
|
| 5 |
"artifact": "docs/data/task_method_20_gap_audit.json",
|
| 6 |
"id": "gap_audit",
|
| 7 |
+
"purpose": "Keep the 61 scoreless cells visible and reproducible."
|
| 8 |
},
|
| 9 |
{
|
| 10 |
"artifact": "scripts/omni/score_model_output_probes.py",
|
|
|
|
| 101 |
"proxy_scored_task_count": 0,
|
| 102 |
"result_record_count": 20,
|
| 103 |
"scope": "128 selected episodes, held-out test",
|
| 104 |
+
"scored_task_count": 13,
|
| 105 |
+
"scoreless_task_count": 7,
|
| 106 |
"status_counts": {
|
| 107 |
+
"not_evaluated_in_verified_package": 7,
|
| 108 |
+
"scored": 13
|
| 109 |
}
|
| 110 |
},
|
| 111 |
"raw128_neural_mlp": {
|
|
|
|
| 140 |
"cosmos3_super_reasoner": 13,
|
| 141 |
"metadata128_neural_mlp": 14,
|
| 142 |
"metadata128_simple": 12,
|
| 143 |
+
"qwen3_omni_v6_lora": 7
|
| 144 |
},
|
| 145 |
"missing_by_status": {
|
| 146 |
+
"not_evaluated_in_verified_package": 35,
|
| 147 |
"not_supported_by_metadata_only_package": 22,
|
| 148 |
"unsupported_without_required_target": 4
|
| 149 |
},
|
|
|
|
| 183 |
"11 Temporal Order Verification": [
|
| 184 |
"cosmos3_nano_future_window",
|
| 185 |
"cosmos3_super_reasoner",
|
| 186 |
+
"metadata128_neural_mlp"
|
|
|
|
| 187 |
],
|
| 188 |
"12 Multimodal Synchronization Detection": [
|
| 189 |
"cosmos3_nano_future_window",
|
| 190 |
"cosmos3_super_reasoner",
|
| 191 |
"metadata128_neural_mlp",
|
| 192 |
+
"metadata128_simple"
|
|
|
|
| 193 |
],
|
| 194 |
"13 Long-Horizon Next-Action Forecasting": [
|
| 195 |
"cosmos3_nano_future_window",
|
|
|
|
| 239 |
"cosmos3_nano_future_window",
|
| 240 |
"cosmos3_super_reasoner",
|
| 241 |
"metadata128_neural_mlp",
|
| 242 |
+
"metadata128_simple"
|
|
|
|
| 243 |
]
|
| 244 |
},
|
| 245 |
"missing_records": [
|
|
|
|
| 516 |
"task_label": "Temporal Order Verification",
|
| 517 |
"task_number": 11
|
| 518 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 519 |
{
|
| 520 |
"method": "Cosmos3-Super Reasoner",
|
| 521 |
"metric_key": "f1",
|
|
|
|
| 568 |
"task_label": "Multimodal Synchronization Detection",
|
| 569 |
"task_number": 12
|
| 570 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 571 |
{
|
| 572 |
"method": "Cosmos3-Super Reasoner",
|
| 573 |
"metric_key": "f1",
|
|
|
|
| 1010 |
"task_label": "Time-to-Next-Transition Regression",
|
| 1011 |
"task_number": 20
|
| 1012 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1013 |
{
|
| 1014 |
"method": "Cosmos3-Super Reasoner",
|
| 1015 |
"metric_key": "mae",
|
|
|
|
| 1083 |
"method_count": 9,
|
| 1084 |
"method_task_record_count": 180,
|
| 1085 |
"proxy_scored_method_task_count": 4,
|
| 1086 |
+
"scored_method_task_count": 119,
|
| 1087 |
+
"scoreless_method_task_count": 61,
|
| 1088 |
"task_count": 20
|
| 1089 |
},
|
| 1090 |
"source_matrix": "docs/data/task_method_20_result_matrix.json",
|
docs/data/task_method_20_result_matrix.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
"title": "Task Method 20-Result Matrix",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
-
"scored_method_task_count":
|
| 9 |
"series": [
|
| 10 |
{
|
| 11 |
"id": "minimal",
|
|
@@ -161,17 +161,17 @@
|
|
| 161 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 162 |
"plotted_as": "colored point overlay",
|
| 163 |
"result_record_count": 20,
|
| 164 |
-
"scored_task_count":
|
| 165 |
-
"covered_task_count":
|
| 166 |
"proxy_scored_task_count": 0,
|
| 167 |
-
"scoreless_task_count":
|
| 168 |
"unsupported_task_count": 0,
|
| 169 |
-
"not_evaluated_task_count":
|
| 170 |
"status_counts": {
|
| 171 |
-
"not_evaluated_in_verified_package":
|
| 172 |
-
"scored":
|
| 173 |
},
|
| 174 |
-
"coverage_fraction": 0.
|
| 175 |
"result_record_fraction": 1.0
|
| 176 |
},
|
| 177 |
{
|
|
@@ -1958,17 +1958,17 @@
|
|
| 1958 |
"task_label": "Temporal Order Verification",
|
| 1959 |
"series_id": "qwen3_omni_v6_lora",
|
| 1960 |
"method": "Qwen3-Omni v6 LoRA",
|
| 1961 |
-
"status": "
|
| 1962 |
-
"status_label": "
|
| 1963 |
-
"scored":
|
| 1964 |
"proxy_scored": false,
|
| 1965 |
-
"raw":
|
| 1966 |
-
"raw_text": "
|
| 1967 |
-
"normalized_score":
|
| 1968 |
-
"metric_key": "
|
| 1969 |
-
"source":
|
| 1970 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1971 |
-
"reason":
|
| 1972 |
},
|
| 1973 |
{
|
| 1974 |
"task_number": 11,
|
|
@@ -2120,17 +2120,17 @@
|
|
| 2120 |
"task_label": "Multimodal Synchronization Detection",
|
| 2121 |
"series_id": "qwen3_omni_v6_lora",
|
| 2122 |
"method": "Qwen3-Omni v6 LoRA",
|
| 2123 |
-
"status": "
|
| 2124 |
-
"status_label": "
|
| 2125 |
-
"scored":
|
| 2126 |
"proxy_scored": false,
|
| 2127 |
-
"raw":
|
| 2128 |
-
"raw_text": "
|
| 2129 |
-
"normalized_score":
|
| 2130 |
-
"metric_key": "
|
| 2131 |
-
"source":
|
| 2132 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 2133 |
-
"reason":
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
"task_number": 12,
|
|
@@ -3416,17 +3416,17 @@
|
|
| 3416 |
"task_label": "Time-to-Next-Transition Regression",
|
| 3417 |
"series_id": "qwen3_omni_v6_lora",
|
| 3418 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3419 |
-
"status": "
|
| 3420 |
-
"status_label": "
|
| 3421 |
-
"scored":
|
| 3422 |
"proxy_scored": false,
|
| 3423 |
-
"raw":
|
| 3424 |
-
"raw_text": "
|
| 3425 |
-
"normalized_score":
|
| 3426 |
-
"metric_key": "
|
| 3427 |
-
"source":
|
| 3428 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3429 |
-
"reason":
|
| 3430 |
},
|
| 3431 |
{
|
| 3432 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Task Method 20-Result Matrix",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
+
"scored_method_task_count": 119,
|
| 9 |
"series": [
|
| 10 |
{
|
| 11 |
"id": "minimal",
|
|
|
|
| 161 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 162 |
"plotted_as": "colored point overlay",
|
| 163 |
"result_record_count": 20,
|
| 164 |
+
"scored_task_count": 13,
|
| 165 |
+
"covered_task_count": 13,
|
| 166 |
"proxy_scored_task_count": 0,
|
| 167 |
+
"scoreless_task_count": 7,
|
| 168 |
"unsupported_task_count": 0,
|
| 169 |
+
"not_evaluated_task_count": 7,
|
| 170 |
"status_counts": {
|
| 171 |
+
"not_evaluated_in_verified_package": 7,
|
| 172 |
+
"scored": 13
|
| 173 |
},
|
| 174 |
+
"coverage_fraction": 0.65,
|
| 175 |
"result_record_fraction": 1.0
|
| 176 |
},
|
| 177 |
{
|
|
|
|
| 1958 |
"task_label": "Temporal Order Verification",
|
| 1959 |
"series_id": "qwen3_omni_v6_lora",
|
| 1960 |
"method": "Qwen3-Omni v6 LoRA",
|
| 1961 |
+
"status": "scored",
|
| 1962 |
+
"status_label": "scored",
|
| 1963 |
+
"scored": true,
|
| 1964 |
"proxy_scored": false,
|
| 1965 |
+
"raw": 0.40984631701404173,
|
| 1966 |
+
"raw_text": "0.4098",
|
| 1967 |
+
"normalized_score": 0.40984631701404173,
|
| 1968 |
+
"metric_key": "temporal_order_f1",
|
| 1969 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1970 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1971 |
+
"reason": null
|
| 1972 |
},
|
| 1973 |
{
|
| 1974 |
"task_number": 11,
|
|
|
|
| 2120 |
"task_label": "Multimodal Synchronization Detection",
|
| 2121 |
"series_id": "qwen3_omni_v6_lora",
|
| 2122 |
"method": "Qwen3-Omni v6 LoRA",
|
| 2123 |
+
"status": "scored",
|
| 2124 |
+
"status_label": "scored",
|
| 2125 |
+
"scored": true,
|
| 2126 |
"proxy_scored": false,
|
| 2127 |
+
"raw": 0.3344936184319576,
|
| 2128 |
+
"raw_text": "0.3345",
|
| 2129 |
+
"normalized_score": 0.3344936184319576,
|
| 2130 |
+
"metric_key": "misalignment_detection_f1",
|
| 2131 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 2132 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 2133 |
+
"reason": null
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
"task_number": 12,
|
|
|
|
| 3416 |
"task_label": "Time-to-Next-Transition Regression",
|
| 3417 |
"series_id": "qwen3_omni_v6_lora",
|
| 3418 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3419 |
+
"status": "scored",
|
| 3420 |
+
"status_label": "scored",
|
| 3421 |
+
"scored": true,
|
| 3422 |
"proxy_scored": false,
|
| 3423 |
+
"raw": 134.0687422166874,
|
| 3424 |
+
"raw_text": "134.07",
|
| 3425 |
+
"normalized_score": 0.07859666766782253,
|
| 3426 |
+
"metric_key": "time_to_transition_mae",
|
| 3427 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 3428 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3429 |
+
"reason": null
|
| 3430 |
},
|
| 3431 |
{
|
| 3432 |
"task_number": 20,
|
docs/data/task_surface_integrity.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-
|
| 4 |
"summary": {
|
| 5 |
"task_count": 12,
|
| 6 |
"expected_task_count": 12,
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:26+00:00",
|
| 4 |
"summary": {
|
| 5 |
"task_count": 12,
|
| 6 |
"expected_task_count": 12,
|
docs/data/unified_task_model_radar.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
"title": "Unified 20-Task Model Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
-
"scored_method_task_count":
|
| 9 |
"normalization_policy": {
|
| 10 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 11 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
@@ -170,17 +170,17 @@
|
|
| 170 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 171 |
"plotted_as": "colored point overlay",
|
| 172 |
"result_record_count": 20,
|
| 173 |
-
"scored_task_count":
|
| 174 |
-
"covered_task_count":
|
| 175 |
"proxy_scored_task_count": 0,
|
| 176 |
-
"scoreless_task_count":
|
| 177 |
"unsupported_task_count": 0,
|
| 178 |
-
"not_evaluated_task_count":
|
| 179 |
"status_counts": {
|
| 180 |
-
"not_evaluated_in_verified_package":
|
| 181 |
-
"scored":
|
| 182 |
},
|
| 183 |
-
"coverage_fraction": 0.
|
| 184 |
"result_record_fraction": 1.0
|
| 185 |
},
|
| 186 |
{
|
|
@@ -1375,6 +1375,17 @@
|
|
| 1375 |
"raw_text": "0.8520",
|
| 1376 |
"status_label": "scored"
|
| 1377 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1378 |
"metadata128_simple": {
|
| 1379 |
"raw": 0.4198864140782312,
|
| 1380 |
"metric_key": "f1",
|
|
@@ -1419,17 +1430,6 @@
|
|
| 1419 |
"raw_text": "n/a",
|
| 1420 |
"status_label": "not supported"
|
| 1421 |
},
|
| 1422 |
-
"qwen3_omni_v6_lora": {
|
| 1423 |
-
"raw": null,
|
| 1424 |
-
"metric_key": "f1",
|
| 1425 |
-
"source": null,
|
| 1426 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1427 |
-
"status": "not_evaluated_in_verified_package",
|
| 1428 |
-
"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",
|
| 1429 |
-
"normalized_score": null,
|
| 1430 |
-
"raw_text": "n/a",
|
| 1431 |
-
"status_label": "not evaluated"
|
| 1432 |
-
},
|
| 1433 |
"cosmos3_super_reasoner": {
|
| 1434 |
"raw": null,
|
| 1435 |
"metric_key": "f1",
|
|
@@ -1486,6 +1486,17 @@
|
|
| 1486 |
"raw_text": "0.7153",
|
| 1487 |
"status_label": "scored"
|
| 1488 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1489 |
"metadata128_simple": {
|
| 1490 |
"raw": null,
|
| 1491 |
"metric_key": "f1",
|
|
@@ -1530,17 +1541,6 @@
|
|
| 1530 |
"raw_text": "n/a",
|
| 1531 |
"status_label": "not supported"
|
| 1532 |
},
|
| 1533 |
-
"qwen3_omni_v6_lora": {
|
| 1534 |
-
"raw": null,
|
| 1535 |
-
"metric_key": "f1",
|
| 1536 |
-
"source": null,
|
| 1537 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1538 |
-
"status": "not_evaluated_in_verified_package",
|
| 1539 |
-
"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",
|
| 1540 |
-
"normalized_score": null,
|
| 1541 |
-
"raw_text": "n/a",
|
| 1542 |
-
"status_label": "not evaluated"
|
| 1543 |
-
},
|
| 1544 |
"cosmos3_super_reasoner": {
|
| 1545 |
"raw": null,
|
| 1546 |
"metric_key": "f1",
|
|
@@ -2374,6 +2374,17 @@
|
|
| 2374 |
"raw_text": "10.55",
|
| 2375 |
"status_label": "scored"
|
| 2376 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2377 |
"raw128_simple": {
|
| 2378 |
"raw": 52.32759475708008,
|
| 2379 |
"metric_key": "mae",
|
|
@@ -2418,17 +2429,6 @@
|
|
| 2418 |
"raw_text": "n/a",
|
| 2419 |
"status_label": "not supported"
|
| 2420 |
},
|
| 2421 |
-
"qwen3_omni_v6_lora": {
|
| 2422 |
-
"raw": null,
|
| 2423 |
-
"metric_key": "mae",
|
| 2424 |
-
"source": null,
|
| 2425 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 2426 |
-
"status": "not_evaluated_in_verified_package",
|
| 2427 |
-
"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",
|
| 2428 |
-
"normalized_score": null,
|
| 2429 |
-
"raw_text": "n/a",
|
| 2430 |
-
"status_label": "not evaluated"
|
| 2431 |
-
},
|
| 2432 |
"cosmos3_super_reasoner": {
|
| 2433 |
"raw": null,
|
| 2434 |
"metric_key": "mae",
|
|
@@ -2492,7 +2492,7 @@
|
|
| 2492 |
"title": "Qwen3-Omni v6 LoRA",
|
| 2493 |
"status": "verified",
|
| 2494 |
"task_aligned_axes": "Qwen3",
|
| 2495 |
-
"coverage": "20 records /
|
| 2496 |
"headline": "JSON validity 0.9990; action macro-F1 0.0029",
|
| 2497 |
"source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_multiscale_cap96_v6_rank64_lr5e5_full8gpu_lora_eval_test_full/eval/metrics.json"
|
| 2498 |
},
|
|
@@ -4256,17 +4256,17 @@
|
|
| 4256 |
"task_label": "Temporal Order Verification",
|
| 4257 |
"series_id": "qwen3_omni_v6_lora",
|
| 4258 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4259 |
-
"status": "
|
| 4260 |
-
"status_label": "
|
| 4261 |
-
"scored":
|
| 4262 |
"proxy_scored": false,
|
| 4263 |
-
"raw":
|
| 4264 |
-
"raw_text": "
|
| 4265 |
-
"normalized_score":
|
| 4266 |
-
"metric_key": "
|
| 4267 |
-
"source":
|
| 4268 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4269 |
-
"reason":
|
| 4270 |
},
|
| 4271 |
{
|
| 4272 |
"task_number": 11,
|
|
@@ -4418,17 +4418,17 @@
|
|
| 4418 |
"task_label": "Multimodal Synchronization Detection",
|
| 4419 |
"series_id": "qwen3_omni_v6_lora",
|
| 4420 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4421 |
-
"status": "
|
| 4422 |
-
"status_label": "
|
| 4423 |
-
"scored":
|
| 4424 |
"proxy_scored": false,
|
| 4425 |
-
"raw":
|
| 4426 |
-
"raw_text": "
|
| 4427 |
-
"normalized_score":
|
| 4428 |
-
"metric_key": "
|
| 4429 |
-
"source":
|
| 4430 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4431 |
-
"reason":
|
| 4432 |
},
|
| 4433 |
{
|
| 4434 |
"task_number": 12,
|
|
@@ -5714,17 +5714,17 @@
|
|
| 5714 |
"task_label": "Time-to-Next-Transition Regression",
|
| 5715 |
"series_id": "qwen3_omni_v6_lora",
|
| 5716 |
"method": "Qwen3-Omni v6 LoRA",
|
| 5717 |
-
"status": "
|
| 5718 |
-
"status_label": "
|
| 5719 |
-
"scored":
|
| 5720 |
"proxy_scored": false,
|
| 5721 |
-
"raw":
|
| 5722 |
-
"raw_text": "
|
| 5723 |
-
"normalized_score":
|
| 5724 |
-
"metric_key": "
|
| 5725 |
-
"source":
|
| 5726 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 5727 |
-
"reason":
|
| 5728 |
},
|
| 5729 |
{
|
| 5730 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Unified 20-Task Model Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
+
"scored_method_task_count": 119,
|
| 9 |
"normalization_policy": {
|
| 10 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 11 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
|
|
| 170 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 171 |
"plotted_as": "colored point overlay",
|
| 172 |
"result_record_count": 20,
|
| 173 |
+
"scored_task_count": 13,
|
| 174 |
+
"covered_task_count": 13,
|
| 175 |
"proxy_scored_task_count": 0,
|
| 176 |
+
"scoreless_task_count": 7,
|
| 177 |
"unsupported_task_count": 0,
|
| 178 |
+
"not_evaluated_task_count": 7,
|
| 179 |
"status_counts": {
|
| 180 |
+
"not_evaluated_in_verified_package": 7,
|
| 181 |
+
"scored": 13
|
| 182 |
},
|
| 183 |
+
"coverage_fraction": 0.65,
|
| 184 |
"result_record_fraction": 1.0
|
| 185 |
},
|
| 186 |
{
|
|
|
|
| 1375 |
"raw_text": "0.8520",
|
| 1376 |
"status_label": "scored"
|
| 1377 |
},
|
| 1378 |
+
"qwen3_omni_v6_lora": {
|
| 1379 |
+
"raw": 0.40984631701404173,
|
| 1380 |
+
"metric_key": "temporal_order_f1",
|
| 1381 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1382 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 1383 |
+
"status": "scored",
|
| 1384 |
+
"reason": null,
|
| 1385 |
+
"normalized_score": 0.40984631701404173,
|
| 1386 |
+
"raw_text": "0.4098",
|
| 1387 |
+
"status_label": "scored"
|
| 1388 |
+
},
|
| 1389 |
"metadata128_simple": {
|
| 1390 |
"raw": 0.4198864140782312,
|
| 1391 |
"metric_key": "f1",
|
|
|
|
| 1430 |
"raw_text": "n/a",
|
| 1431 |
"status_label": "not supported"
|
| 1432 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1433 |
"cosmos3_super_reasoner": {
|
| 1434 |
"raw": null,
|
| 1435 |
"metric_key": "f1",
|
|
|
|
| 1486 |
"raw_text": "0.7153",
|
| 1487 |
"status_label": "scored"
|
| 1488 |
},
|
| 1489 |
+
"qwen3_omni_v6_lora": {
|
| 1490 |
+
"raw": 0.3344936184319576,
|
| 1491 |
+
"metric_key": "misalignment_detection_f1",
|
| 1492 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 1493 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 1494 |
+
"status": "scored",
|
| 1495 |
+
"reason": null,
|
| 1496 |
+
"normalized_score": 0.3344936184319576,
|
| 1497 |
+
"raw_text": "0.3345",
|
| 1498 |
+
"status_label": "scored"
|
| 1499 |
+
},
|
| 1500 |
"metadata128_simple": {
|
| 1501 |
"raw": null,
|
| 1502 |
"metric_key": "f1",
|
|
|
|
| 1541 |
"raw_text": "n/a",
|
| 1542 |
"status_label": "not supported"
|
| 1543 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1544 |
"cosmos3_super_reasoner": {
|
| 1545 |
"raw": null,
|
| 1546 |
"metric_key": "f1",
|
|
|
|
| 2374 |
"raw_text": "10.55",
|
| 2375 |
"status_label": "scored"
|
| 2376 |
},
|
| 2377 |
+
"qwen3_omni_v6_lora": {
|
| 2378 |
+
"raw": 134.0687422166874,
|
| 2379 |
+
"metric_key": "time_to_transition_mae",
|
| 2380 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 2381 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 2382 |
+
"status": "scored",
|
| 2383 |
+
"reason": null,
|
| 2384 |
+
"normalized_score": 0.07859666766782253,
|
| 2385 |
+
"raw_text": "134.07",
|
| 2386 |
+
"status_label": "scored"
|
| 2387 |
+
},
|
| 2388 |
"raw128_simple": {
|
| 2389 |
"raw": 52.32759475708008,
|
| 2390 |
"metric_key": "mae",
|
|
|
|
| 2429 |
"raw_text": "n/a",
|
| 2430 |
"status_label": "not supported"
|
| 2431 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2432 |
"cosmos3_super_reasoner": {
|
| 2433 |
"raw": null,
|
| 2434 |
"metric_key": "mae",
|
|
|
|
| 2492 |
"title": "Qwen3-Omni v6 LoRA",
|
| 2493 |
"status": "verified",
|
| 2494 |
"task_aligned_axes": "Qwen3",
|
| 2495 |
+
"coverage": "20 records / 13 scored task-aligned axes",
|
| 2496 |
"headline": "JSON validity 0.9990; action macro-F1 0.0029",
|
| 2497 |
"source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_multiscale_cap96_v6_rank64_lr5e5_full8gpu_lora_eval_test_full/eval/metrics.json"
|
| 2498 |
},
|
|
|
|
| 4256 |
"task_label": "Temporal Order Verification",
|
| 4257 |
"series_id": "qwen3_omni_v6_lora",
|
| 4258 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4259 |
+
"status": "scored",
|
| 4260 |
+
"status_label": "scored",
|
| 4261 |
+
"scored": true,
|
| 4262 |
"proxy_scored": false,
|
| 4263 |
+
"raw": 0.40984631701404173,
|
| 4264 |
+
"raw_text": "0.4098",
|
| 4265 |
+
"normalized_score": 0.40984631701404173,
|
| 4266 |
+
"metric_key": "temporal_order_f1",
|
| 4267 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 4268 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4269 |
+
"reason": null
|
| 4270 |
},
|
| 4271 |
{
|
| 4272 |
"task_number": 11,
|
|
|
|
| 4418 |
"task_label": "Multimodal Synchronization Detection",
|
| 4419 |
"series_id": "qwen3_omni_v6_lora",
|
| 4420 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4421 |
+
"status": "scored",
|
| 4422 |
+
"status_label": "scored",
|
| 4423 |
+
"scored": true,
|
| 4424 |
"proxy_scored": false,
|
| 4425 |
+
"raw": 0.3344936184319576,
|
| 4426 |
+
"raw_text": "0.3345",
|
| 4427 |
+
"normalized_score": 0.3344936184319576,
|
| 4428 |
+
"metric_key": "misalignment_detection_f1",
|
| 4429 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 4430 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4431 |
+
"reason": null
|
| 4432 |
},
|
| 4433 |
{
|
| 4434 |
"task_number": 12,
|
|
|
|
| 5714 |
"task_label": "Time-to-Next-Transition Regression",
|
| 5715 |
"series_id": "qwen3_omni_v6_lora",
|
| 5716 |
"method": "Qwen3-Omni v6 LoRA",
|
| 5717 |
+
"status": "scored",
|
| 5718 |
+
"status_label": "scored",
|
| 5719 |
+
"scored": true,
|
| 5720 |
"proxy_scored": false,
|
| 5721 |
+
"raw": 134.0687422166874,
|
| 5722 |
+
"raw_text": "134.07",
|
| 5723 |
+
"normalized_score": 0.07859666766782253,
|
| 5724 |
+
"metric_key": "time_to_transition_mae",
|
| 5725 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 5726 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 5727 |
+
"reason": null
|
| 5728 |
},
|
| 5729 |
{
|
| 5730 |
"task_number": 20,
|
docs/data/website_integrity.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-17T21:
|
| 4 |
"docs_root": "docs",
|
| 5 |
"site_base": "/ropedia-xperience-10m-task-suite/",
|
| 6 |
"summary": {
|
|
@@ -316,7 +316,7 @@
|
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"path": "data/episode128_task_model_radar.json",
|
| 319 |
-
"bytes":
|
| 320 |
"top_level_type": "dict"
|
| 321 |
},
|
| 322 |
{
|
|
@@ -486,12 +486,12 @@
|
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"path": "data/task_method_20_gap_audit.json",
|
| 489 |
-
"bytes":
|
| 490 |
"top_level_type": "dict"
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"path": "data/task_method_20_result_matrix.json",
|
| 494 |
-
"bytes":
|
| 495 |
"top_level_type": "dict"
|
| 496 |
},
|
| 497 |
{
|
|
@@ -526,7 +526,7 @@
|
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"path": "data/unified_task_model_radar.json",
|
| 529 |
-
"bytes":
|
| 530 |
"top_level_type": "dict"
|
| 531 |
},
|
| 532 |
{
|
|
@@ -566,7 +566,7 @@
|
|
| 566 |
{
|
| 567 |
"path": "assets/charts/episode128_task_model_radar.svg",
|
| 568 |
"exists": true,
|
| 569 |
-
"bytes":
|
| 570 |
"format": "SVG",
|
| 571 |
"has_viewbox": true
|
| 572 |
},
|
|
@@ -636,7 +636,7 @@
|
|
| 636 |
{
|
| 637 |
"path": "assets/charts/unified_task_model_radar.svg",
|
| 638 |
"exists": true,
|
| 639 |
-
"bytes":
|
| 640 |
"format": "SVG",
|
| 641 |
"has_viewbox": true
|
| 642 |
},
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:27+00:00",
|
| 4 |
"docs_root": "docs",
|
| 5 |
"site_base": "/ropedia-xperience-10m-task-suite/",
|
| 6 |
"summary": {
|
|
|
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"path": "data/episode128_task_model_radar.json",
|
| 319 |
+
"bytes": 187309,
|
| 320 |
"top_level_type": "dict"
|
| 321 |
},
|
| 322 |
{
|
|
|
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"path": "data/task_method_20_gap_audit.json",
|
| 489 |
+
"bytes": 53574,
|
| 490 |
"top_level_type": "dict"
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"path": "data/task_method_20_result_matrix.json",
|
| 494 |
+
"bytes": 129707,
|
| 495 |
"top_level_type": "dict"
|
| 496 |
},
|
| 497 |
{
|
|
|
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"path": "data/unified_task_model_radar.json",
|
| 529 |
+
"bytes": 231161,
|
| 530 |
"top_level_type": "dict"
|
| 531 |
},
|
| 532 |
{
|
|
|
|
| 566 |
{
|
| 567 |
"path": "assets/charts/episode128_task_model_radar.svg",
|
| 568 |
"exists": true,
|
| 569 |
+
"bytes": 44378,
|
| 570 |
"format": "SVG",
|
| 571 |
"has_viewbox": true
|
| 572 |
},
|
|
|
|
| 636 |
{
|
| 637 |
"path": "assets/charts/unified_task_model_radar.svg",
|
| 638 |
"exists": true,
|
| 639 |
+
"bytes": 50394,
|
| 640 |
"format": "SVG",
|
| 641 |
"has_viewbox": true
|
| 642 |
},
|
metrics/episode128_task_model_radar.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"title": "128-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"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.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 7,
|
| 8 |
"method_task_record_count": 140,
|
| 9 |
-
"scored_method_task_count":
|
| 10 |
"normalization_policy": {
|
| 11 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 12 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
@@ -127,17 +127,17 @@
|
|
| 127 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 128 |
"plotted_as": "colored point overlay",
|
| 129 |
"result_record_count": 20,
|
| 130 |
-
"scored_task_count":
|
| 131 |
-
"covered_task_count":
|
| 132 |
"proxy_scored_task_count": 0,
|
| 133 |
-
"scoreless_task_count":
|
| 134 |
"unsupported_task_count": 0,
|
| 135 |
-
"not_evaluated_task_count":
|
| 136 |
"status_counts": {
|
| 137 |
-
"not_evaluated_in_verified_package":
|
| 138 |
-
"scored":
|
| 139 |
},
|
| 140 |
-
"coverage_fraction": 0.
|
| 141 |
"result_record_fraction": 1.0
|
| 142 |
},
|
| 143 |
{
|
|
@@ -1157,15 +1157,15 @@
|
|
| 1157 |
"status_label": "scored"
|
| 1158 |
},
|
| 1159 |
"qwen3_omni_v6_lora": {
|
| 1160 |
-
"raw":
|
| 1161 |
-
"metric_key": "
|
| 1162 |
-
"source":
|
| 1163 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1164 |
-
"status": "
|
| 1165 |
-
"reason":
|
| 1166 |
-
"normalized_score":
|
| 1167 |
-
"raw_text": "
|
| 1168 |
-
"status_label": "
|
| 1169 |
},
|
| 1170 |
"cosmos3_super_reasoner": {
|
| 1171 |
"raw": null,
|
|
@@ -1248,15 +1248,15 @@
|
|
| 1248 |
"status_label": "scored"
|
| 1249 |
},
|
| 1250 |
"qwen3_omni_v6_lora": {
|
| 1251 |
-
"raw":
|
| 1252 |
-
"metric_key": "
|
| 1253 |
-
"source":
|
| 1254 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1255 |
-
"status": "
|
| 1256 |
-
"reason":
|
| 1257 |
-
"normalized_score":
|
| 1258 |
-
"raw_text": "
|
| 1259 |
-
"status_label": "
|
| 1260 |
},
|
| 1261 |
"cosmos3_super_reasoner": {
|
| 1262 |
"raw": null,
|
|
@@ -1976,15 +1976,15 @@
|
|
| 1976 |
"status_label": "scored"
|
| 1977 |
},
|
| 1978 |
"qwen3_omni_v6_lora": {
|
| 1979 |
-
"raw":
|
| 1980 |
-
"metric_key": "
|
| 1981 |
-
"source":
|
| 1982 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1983 |
-
"status": "
|
| 1984 |
-
"reason":
|
| 1985 |
-
"normalized_score":
|
| 1986 |
-
"raw_text": "
|
| 1987 |
-
"status_label": "
|
| 1988 |
},
|
| 1989 |
"cosmos3_super_reasoner": {
|
| 1990 |
"raw": null,
|
|
@@ -3350,17 +3350,17 @@
|
|
| 3350 |
"task_label": "Temporal Order Verification",
|
| 3351 |
"series_id": "qwen3_omni_v6_lora",
|
| 3352 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3353 |
-
"status": "
|
| 3354 |
-
"status_label": "
|
| 3355 |
-
"scored":
|
| 3356 |
"proxy_scored": false,
|
| 3357 |
-
"raw":
|
| 3358 |
-
"raw_text": "
|
| 3359 |
-
"normalized_score":
|
| 3360 |
-
"metric_key": "
|
| 3361 |
-
"source":
|
| 3362 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3363 |
-
"reason":
|
| 3364 |
},
|
| 3365 |
{
|
| 3366 |
"task_number": 11,
|
|
@@ -3476,17 +3476,17 @@
|
|
| 3476 |
"task_label": "Multimodal Synchronization Detection",
|
| 3477 |
"series_id": "qwen3_omni_v6_lora",
|
| 3478 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3479 |
-
"status": "
|
| 3480 |
-
"status_label": "
|
| 3481 |
-
"scored":
|
| 3482 |
"proxy_scored": false,
|
| 3483 |
-
"raw":
|
| 3484 |
-
"raw_text": "
|
| 3485 |
-
"normalized_score":
|
| 3486 |
-
"metric_key": "
|
| 3487 |
-
"source":
|
| 3488 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3489 |
-
"reason":
|
| 3490 |
},
|
| 3491 |
{
|
| 3492 |
"task_number": 12,
|
|
@@ -4484,17 +4484,17 @@
|
|
| 4484 |
"task_label": "Time-to-Next-Transition Regression",
|
| 4485 |
"series_id": "qwen3_omni_v6_lora",
|
| 4486 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4487 |
-
"status": "
|
| 4488 |
-
"status_label": "
|
| 4489 |
-
"scored":
|
| 4490 |
"proxy_scored": false,
|
| 4491 |
-
"raw":
|
| 4492 |
-
"raw_text": "
|
| 4493 |
-
"normalized_score":
|
| 4494 |
-
"metric_key": "
|
| 4495 |
-
"source":
|
| 4496 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4497 |
-
"reason":
|
| 4498 |
},
|
| 4499 |
{
|
| 4500 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "128-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"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.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 7,
|
| 8 |
"method_task_record_count": 140,
|
| 9 |
+
"scored_method_task_count": 79,
|
| 10 |
"normalization_policy": {
|
| 11 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 12 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
|
|
| 127 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 128 |
"plotted_as": "colored point overlay",
|
| 129 |
"result_record_count": 20,
|
| 130 |
+
"scored_task_count": 13,
|
| 131 |
+
"covered_task_count": 13,
|
| 132 |
"proxy_scored_task_count": 0,
|
| 133 |
+
"scoreless_task_count": 7,
|
| 134 |
"unsupported_task_count": 0,
|
| 135 |
+
"not_evaluated_task_count": 7,
|
| 136 |
"status_counts": {
|
| 137 |
+
"not_evaluated_in_verified_package": 7,
|
| 138 |
+
"scored": 13
|
| 139 |
},
|
| 140 |
+
"coverage_fraction": 0.65,
|
| 141 |
"result_record_fraction": 1.0
|
| 142 |
},
|
| 143 |
{
|
|
|
|
| 1157 |
"status_label": "scored"
|
| 1158 |
},
|
| 1159 |
"qwen3_omni_v6_lora": {
|
| 1160 |
+
"raw": 0.40984631701404173,
|
| 1161 |
+
"metric_key": "temporal_order_f1",
|
| 1162 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1163 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1164 |
+
"status": "scored",
|
| 1165 |
+
"reason": null,
|
| 1166 |
+
"normalized_score": 0.40984631701404173,
|
| 1167 |
+
"raw_text": "0.4098",
|
| 1168 |
+
"status_label": "scored"
|
| 1169 |
},
|
| 1170 |
"cosmos3_super_reasoner": {
|
| 1171 |
"raw": null,
|
|
|
|
| 1248 |
"status_label": "scored"
|
| 1249 |
},
|
| 1250 |
"qwen3_omni_v6_lora": {
|
| 1251 |
+
"raw": 0.3344936184319576,
|
| 1252 |
+
"metric_key": "misalignment_detection_f1",
|
| 1253 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 1254 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1255 |
+
"status": "scored",
|
| 1256 |
+
"reason": null,
|
| 1257 |
+
"normalized_score": 0.3344936184319576,
|
| 1258 |
+
"raw_text": "0.3345",
|
| 1259 |
+
"status_label": "scored"
|
| 1260 |
},
|
| 1261 |
"cosmos3_super_reasoner": {
|
| 1262 |
"raw": null,
|
|
|
|
| 1976 |
"status_label": "scored"
|
| 1977 |
},
|
| 1978 |
"qwen3_omni_v6_lora": {
|
| 1979 |
+
"raw": 134.0687422166874,
|
| 1980 |
+
"metric_key": "time_to_transition_mae",
|
| 1981 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 1982 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1983 |
+
"status": "scored",
|
| 1984 |
+
"reason": null,
|
| 1985 |
+
"normalized_score": 0.07859666766782253,
|
| 1986 |
+
"raw_text": "134.07",
|
| 1987 |
+
"status_label": "scored"
|
| 1988 |
},
|
| 1989 |
"cosmos3_super_reasoner": {
|
| 1990 |
"raw": null,
|
|
|
|
| 3350 |
"task_label": "Temporal Order Verification",
|
| 3351 |
"series_id": "qwen3_omni_v6_lora",
|
| 3352 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3353 |
+
"status": "scored",
|
| 3354 |
+
"status_label": "scored",
|
| 3355 |
+
"scored": true,
|
| 3356 |
"proxy_scored": false,
|
| 3357 |
+
"raw": 0.40984631701404173,
|
| 3358 |
+
"raw_text": "0.4098",
|
| 3359 |
+
"normalized_score": 0.40984631701404173,
|
| 3360 |
+
"metric_key": "temporal_order_f1",
|
| 3361 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 3362 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3363 |
+
"reason": null
|
| 3364 |
},
|
| 3365 |
{
|
| 3366 |
"task_number": 11,
|
|
|
|
| 3476 |
"task_label": "Multimodal Synchronization Detection",
|
| 3477 |
"series_id": "qwen3_omni_v6_lora",
|
| 3478 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3479 |
+
"status": "scored",
|
| 3480 |
+
"status_label": "scored",
|
| 3481 |
+
"scored": true,
|
| 3482 |
"proxy_scored": false,
|
| 3483 |
+
"raw": 0.3344936184319576,
|
| 3484 |
+
"raw_text": "0.3345",
|
| 3485 |
+
"normalized_score": 0.3344936184319576,
|
| 3486 |
+
"metric_key": "misalignment_detection_f1",
|
| 3487 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 3488 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3489 |
+
"reason": null
|
| 3490 |
},
|
| 3491 |
{
|
| 3492 |
"task_number": 12,
|
|
|
|
| 4484 |
"task_label": "Time-to-Next-Transition Regression",
|
| 4485 |
"series_id": "qwen3_omni_v6_lora",
|
| 4486 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4487 |
+
"status": "scored",
|
| 4488 |
+
"status_label": "scored",
|
| 4489 |
+
"scored": true,
|
| 4490 |
"proxy_scored": false,
|
| 4491 |
+
"raw": 134.0687422166874,
|
| 4492 |
+
"raw_text": "134.07",
|
| 4493 |
+
"normalized_score": 0.07859666766782253,
|
| 4494 |
+
"metric_key": "time_to_transition_mae",
|
| 4495 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 4496 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4497 |
+
"reason": null
|
| 4498 |
},
|
| 4499 |
{
|
| 4500 |
"task_number": 20,
|
metrics/publication_audit.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-17T21:
|
| 4 |
"checks": [
|
| 5 |
{
|
| 6 |
"name": "required_publication_assets_present",
|
|
@@ -206,8 +206,8 @@
|
|
| 206 |
"github_repo": {
|
| 207 |
"root": "repo",
|
| 208 |
"exists": true,
|
| 209 |
-
"file_count":
|
| 210 |
-
"text_file_count":
|
| 211 |
"largest_file": {
|
| 212 |
"path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
|
| 213 |
"bytes": 55702978
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:35+00:00",
|
| 4 |
"checks": [
|
| 5 |
{
|
| 6 |
"name": "required_publication_assets_present",
|
|
|
|
| 206 |
"github_repo": {
|
| 207 |
"root": "repo",
|
| 208 |
"exists": true,
|
| 209 |
+
"file_count": 1250,
|
| 210 |
+
"text_file_count": 1052,
|
| 211 |
"largest_file": {
|
| 212 |
"path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
|
| 213 |
"bytes": 55702978
|
metrics/single_episode_task_model_radar.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"title": "Single-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"description": "Minimal and Neural MLP baselines on the one public sample episode, both scored on all 20 task contracts.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 2,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Single-Episode 20-Task Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"description": "Minimal and Neural MLP baselines on the one public sample episode, both scored on all 20 task contracts.",
|
| 6 |
"task_count": 20,
|
| 7 |
"method_count": 2,
|
metrics/task_method_20_gap_audit.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"generated_at_utc": "2026-06-
|
| 3 |
"immediate_actions": [
|
| 4 |
{
|
| 5 |
"artifact": "docs/data/task_method_20_gap_audit.json",
|
| 6 |
"id": "gap_audit",
|
| 7 |
-
"purpose": "Keep the
|
| 8 |
},
|
| 9 |
{
|
| 10 |
"artifact": "scripts/omni/score_model_output_probes.py",
|
|
@@ -101,11 +101,11 @@
|
|
| 101 |
"proxy_scored_task_count": 0,
|
| 102 |
"result_record_count": 20,
|
| 103 |
"scope": "128 selected episodes, held-out test",
|
| 104 |
-
"scored_task_count":
|
| 105 |
-
"scoreless_task_count":
|
| 106 |
"status_counts": {
|
| 107 |
-
"not_evaluated_in_verified_package":
|
| 108 |
-
"scored":
|
| 109 |
}
|
| 110 |
},
|
| 111 |
"raw128_neural_mlp": {
|
|
@@ -140,10 +140,10 @@
|
|
| 140 |
"cosmos3_super_reasoner": 13,
|
| 141 |
"metadata128_neural_mlp": 14,
|
| 142 |
"metadata128_simple": 12,
|
| 143 |
-
"qwen3_omni_v6_lora":
|
| 144 |
},
|
| 145 |
"missing_by_status": {
|
| 146 |
-
"not_evaluated_in_verified_package":
|
| 147 |
"not_supported_by_metadata_only_package": 22,
|
| 148 |
"unsupported_without_required_target": 4
|
| 149 |
},
|
|
@@ -183,15 +183,13 @@
|
|
| 183 |
"11 Temporal Order Verification": [
|
| 184 |
"cosmos3_nano_future_window",
|
| 185 |
"cosmos3_super_reasoner",
|
| 186 |
-
"metadata128_neural_mlp"
|
| 187 |
-
"qwen3_omni_v6_lora"
|
| 188 |
],
|
| 189 |
"12 Multimodal Synchronization Detection": [
|
| 190 |
"cosmos3_nano_future_window",
|
| 191 |
"cosmos3_super_reasoner",
|
| 192 |
"metadata128_neural_mlp",
|
| 193 |
-
"metadata128_simple"
|
| 194 |
-
"qwen3_omni_v6_lora"
|
| 195 |
],
|
| 196 |
"13 Long-Horizon Next-Action Forecasting": [
|
| 197 |
"cosmos3_nano_future_window",
|
|
@@ -241,8 +239,7 @@
|
|
| 241 |
"cosmos3_nano_future_window",
|
| 242 |
"cosmos3_super_reasoner",
|
| 243 |
"metadata128_neural_mlp",
|
| 244 |
-
"metadata128_simple"
|
| 245 |
-
"qwen3_omni_v6_lora"
|
| 246 |
]
|
| 247 |
},
|
| 248 |
"missing_records": [
|
|
@@ -519,19 +516,6 @@
|
|
| 519 |
"task_label": "Temporal Order Verification",
|
| 520 |
"task_number": 11
|
| 521 |
},
|
| 522 |
-
{
|
| 523 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 524 |
-
"metric_key": "f1",
|
| 525 |
-
"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",
|
| 526 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 527 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 528 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 529 |
-
"status": "not_evaluated_in_verified_package",
|
| 530 |
-
"status_label": "not evaluated",
|
| 531 |
-
"task_id": "temporal_order",
|
| 532 |
-
"task_label": "Temporal Order Verification",
|
| 533 |
-
"task_number": 11
|
| 534 |
-
},
|
| 535 |
{
|
| 536 |
"method": "Cosmos3-Super Reasoner",
|
| 537 |
"metric_key": "f1",
|
|
@@ -584,19 +568,6 @@
|
|
| 584 |
"task_label": "Multimodal Synchronization Detection",
|
| 585 |
"task_number": 12
|
| 586 |
},
|
| 587 |
-
{
|
| 588 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 589 |
-
"metric_key": "f1",
|
| 590 |
-
"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",
|
| 591 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 592 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 593 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 594 |
-
"status": "not_evaluated_in_verified_package",
|
| 595 |
-
"status_label": "not evaluated",
|
| 596 |
-
"task_id": "misalignment_detection",
|
| 597 |
-
"task_label": "Multimodal Synchronization Detection",
|
| 598 |
-
"task_number": 12
|
| 599 |
-
},
|
| 600 |
{
|
| 601 |
"method": "Cosmos3-Super Reasoner",
|
| 602 |
"metric_key": "f1",
|
|
@@ -1039,19 +1010,6 @@
|
|
| 1039 |
"task_label": "Time-to-Next-Transition Regression",
|
| 1040 |
"task_number": 20
|
| 1041 |
},
|
| 1042 |
-
{
|
| 1043 |
-
"method": "Qwen3-Omni v6 LoRA",
|
| 1044 |
-
"metric_key": "mae",
|
| 1045 |
-
"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",
|
| 1046 |
-
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
|
| 1047 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1048 |
-
"series_id": "qwen3_omni_v6_lora",
|
| 1049 |
-
"status": "not_evaluated_in_verified_package",
|
| 1050 |
-
"status_label": "not evaluated",
|
| 1051 |
-
"task_id": "time_to_transition",
|
| 1052 |
-
"task_label": "Time-to-Next-Transition Regression",
|
| 1053 |
-
"task_number": 20
|
| 1054 |
-
},
|
| 1055 |
{
|
| 1056 |
"method": "Cosmos3-Super Reasoner",
|
| 1057 |
"metric_key": "mae",
|
|
@@ -1125,8 +1083,8 @@
|
|
| 1125 |
"method_count": 9,
|
| 1126 |
"method_task_record_count": 180,
|
| 1127 |
"proxy_scored_method_task_count": 4,
|
| 1128 |
-
"scored_method_task_count":
|
| 1129 |
-
"scoreless_method_task_count":
|
| 1130 |
"task_count": 20
|
| 1131 |
},
|
| 1132 |
"source_matrix": "docs/data/task_method_20_result_matrix.json",
|
|
|
|
| 1 |
{
|
| 2 |
+
"generated_at_utc": "2026-06-17T21:17:51+00:00",
|
| 3 |
"immediate_actions": [
|
| 4 |
{
|
| 5 |
"artifact": "docs/data/task_method_20_gap_audit.json",
|
| 6 |
"id": "gap_audit",
|
| 7 |
+
"purpose": "Keep the 61 scoreless cells visible and reproducible."
|
| 8 |
},
|
| 9 |
{
|
| 10 |
"artifact": "scripts/omni/score_model_output_probes.py",
|
|
|
|
| 101 |
"proxy_scored_task_count": 0,
|
| 102 |
"result_record_count": 20,
|
| 103 |
"scope": "128 selected episodes, held-out test",
|
| 104 |
+
"scored_task_count": 13,
|
| 105 |
+
"scoreless_task_count": 7,
|
| 106 |
"status_counts": {
|
| 107 |
+
"not_evaluated_in_verified_package": 7,
|
| 108 |
+
"scored": 13
|
| 109 |
}
|
| 110 |
},
|
| 111 |
"raw128_neural_mlp": {
|
|
|
|
| 140 |
"cosmos3_super_reasoner": 13,
|
| 141 |
"metadata128_neural_mlp": 14,
|
| 142 |
"metadata128_simple": 12,
|
| 143 |
+
"qwen3_omni_v6_lora": 7
|
| 144 |
},
|
| 145 |
"missing_by_status": {
|
| 146 |
+
"not_evaluated_in_verified_package": 35,
|
| 147 |
"not_supported_by_metadata_only_package": 22,
|
| 148 |
"unsupported_without_required_target": 4
|
| 149 |
},
|
|
|
|
| 183 |
"11 Temporal Order Verification": [
|
| 184 |
"cosmos3_nano_future_window",
|
| 185 |
"cosmos3_super_reasoner",
|
| 186 |
+
"metadata128_neural_mlp"
|
|
|
|
| 187 |
],
|
| 188 |
"12 Multimodal Synchronization Detection": [
|
| 189 |
"cosmos3_nano_future_window",
|
| 190 |
"cosmos3_super_reasoner",
|
| 191 |
"metadata128_neural_mlp",
|
| 192 |
+
"metadata128_simple"
|
|
|
|
| 193 |
],
|
| 194 |
"13 Long-Horizon Next-Action Forecasting": [
|
| 195 |
"cosmos3_nano_future_window",
|
|
|
|
| 239 |
"cosmos3_nano_future_window",
|
| 240 |
"cosmos3_super_reasoner",
|
| 241 |
"metadata128_neural_mlp",
|
| 242 |
+
"metadata128_simple"
|
|
|
|
| 243 |
]
|
| 244 |
},
|
| 245 |
"missing_records": [
|
|
|
|
| 516 |
"task_label": "Temporal Order Verification",
|
| 517 |
"task_number": 11
|
| 518 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 519 |
{
|
| 520 |
"method": "Cosmos3-Super Reasoner",
|
| 521 |
"metric_key": "f1",
|
|
|
|
| 568 |
"task_label": "Multimodal Synchronization Detection",
|
| 569 |
"task_number": 12
|
| 570 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 571 |
{
|
| 572 |
"method": "Cosmos3-Super Reasoner",
|
| 573 |
"metric_key": "f1",
|
|
|
|
| 1010 |
"task_label": "Time-to-Next-Transition Regression",
|
| 1011 |
"task_number": 20
|
| 1012 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1013 |
{
|
| 1014 |
"method": "Cosmos3-Super Reasoner",
|
| 1015 |
"metric_key": "mae",
|
|
|
|
| 1083 |
"method_count": 9,
|
| 1084 |
"method_task_record_count": 180,
|
| 1085 |
"proxy_scored_method_task_count": 4,
|
| 1086 |
+
"scored_method_task_count": 119,
|
| 1087 |
+
"scoreless_method_task_count": 61,
|
| 1088 |
"task_count": 20
|
| 1089 |
},
|
| 1090 |
"source_matrix": "docs/data/task_method_20_result_matrix.json",
|
metrics/task_method_20_result_matrix.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
"title": "Task Method 20-Result Matrix",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
-
"scored_method_task_count":
|
| 9 |
"series": [
|
| 10 |
{
|
| 11 |
"id": "minimal",
|
|
@@ -161,17 +161,17 @@
|
|
| 161 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 162 |
"plotted_as": "colored point overlay",
|
| 163 |
"result_record_count": 20,
|
| 164 |
-
"scored_task_count":
|
| 165 |
-
"covered_task_count":
|
| 166 |
"proxy_scored_task_count": 0,
|
| 167 |
-
"scoreless_task_count":
|
| 168 |
"unsupported_task_count": 0,
|
| 169 |
-
"not_evaluated_task_count":
|
| 170 |
"status_counts": {
|
| 171 |
-
"not_evaluated_in_verified_package":
|
| 172 |
-
"scored":
|
| 173 |
},
|
| 174 |
-
"coverage_fraction": 0.
|
| 175 |
"result_record_fraction": 1.0
|
| 176 |
},
|
| 177 |
{
|
|
@@ -1958,17 +1958,17 @@
|
|
| 1958 |
"task_label": "Temporal Order Verification",
|
| 1959 |
"series_id": "qwen3_omni_v6_lora",
|
| 1960 |
"method": "Qwen3-Omni v6 LoRA",
|
| 1961 |
-
"status": "
|
| 1962 |
-
"status_label": "
|
| 1963 |
-
"scored":
|
| 1964 |
"proxy_scored": false,
|
| 1965 |
-
"raw":
|
| 1966 |
-
"raw_text": "
|
| 1967 |
-
"normalized_score":
|
| 1968 |
-
"metric_key": "
|
| 1969 |
-
"source":
|
| 1970 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1971 |
-
"reason":
|
| 1972 |
},
|
| 1973 |
{
|
| 1974 |
"task_number": 11,
|
|
@@ -2120,17 +2120,17 @@
|
|
| 2120 |
"task_label": "Multimodal Synchronization Detection",
|
| 2121 |
"series_id": "qwen3_omni_v6_lora",
|
| 2122 |
"method": "Qwen3-Omni v6 LoRA",
|
| 2123 |
-
"status": "
|
| 2124 |
-
"status_label": "
|
| 2125 |
-
"scored":
|
| 2126 |
"proxy_scored": false,
|
| 2127 |
-
"raw":
|
| 2128 |
-
"raw_text": "
|
| 2129 |
-
"normalized_score":
|
| 2130 |
-
"metric_key": "
|
| 2131 |
-
"source":
|
| 2132 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 2133 |
-
"reason":
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
"task_number": 12,
|
|
@@ -3416,17 +3416,17 @@
|
|
| 3416 |
"task_label": "Time-to-Next-Transition Regression",
|
| 3417 |
"series_id": "qwen3_omni_v6_lora",
|
| 3418 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3419 |
-
"status": "
|
| 3420 |
-
"status_label": "
|
| 3421 |
-
"scored":
|
| 3422 |
"proxy_scored": false,
|
| 3423 |
-
"raw":
|
| 3424 |
-
"raw_text": "
|
| 3425 |
-
"normalized_score":
|
| 3426 |
-
"metric_key": "
|
| 3427 |
-
"source":
|
| 3428 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3429 |
-
"reason":
|
| 3430 |
},
|
| 3431 |
{
|
| 3432 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Task Method 20-Result Matrix",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
+
"scored_method_task_count": 119,
|
| 9 |
"series": [
|
| 10 |
{
|
| 11 |
"id": "minimal",
|
|
|
|
| 161 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 162 |
"plotted_as": "colored point overlay",
|
| 163 |
"result_record_count": 20,
|
| 164 |
+
"scored_task_count": 13,
|
| 165 |
+
"covered_task_count": 13,
|
| 166 |
"proxy_scored_task_count": 0,
|
| 167 |
+
"scoreless_task_count": 7,
|
| 168 |
"unsupported_task_count": 0,
|
| 169 |
+
"not_evaluated_task_count": 7,
|
| 170 |
"status_counts": {
|
| 171 |
+
"not_evaluated_in_verified_package": 7,
|
| 172 |
+
"scored": 13
|
| 173 |
},
|
| 174 |
+
"coverage_fraction": 0.65,
|
| 175 |
"result_record_fraction": 1.0
|
| 176 |
},
|
| 177 |
{
|
|
|
|
| 1958 |
"task_label": "Temporal Order Verification",
|
| 1959 |
"series_id": "qwen3_omni_v6_lora",
|
| 1960 |
"method": "Qwen3-Omni v6 LoRA",
|
| 1961 |
+
"status": "scored",
|
| 1962 |
+
"status_label": "scored",
|
| 1963 |
+
"scored": true,
|
| 1964 |
"proxy_scored": false,
|
| 1965 |
+
"raw": 0.40984631701404173,
|
| 1966 |
+
"raw_text": "0.4098",
|
| 1967 |
+
"normalized_score": 0.40984631701404173,
|
| 1968 |
+
"metric_key": "temporal_order_f1",
|
| 1969 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1970 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 1971 |
+
"reason": null
|
| 1972 |
},
|
| 1973 |
{
|
| 1974 |
"task_number": 11,
|
|
|
|
| 2120 |
"task_label": "Multimodal Synchronization Detection",
|
| 2121 |
"series_id": "qwen3_omni_v6_lora",
|
| 2122 |
"method": "Qwen3-Omni v6 LoRA",
|
| 2123 |
+
"status": "scored",
|
| 2124 |
+
"status_label": "scored",
|
| 2125 |
+
"scored": true,
|
| 2126 |
"proxy_scored": false,
|
| 2127 |
+
"raw": 0.3344936184319576,
|
| 2128 |
+
"raw_text": "0.3345",
|
| 2129 |
+
"normalized_score": 0.3344936184319576,
|
| 2130 |
+
"metric_key": "misalignment_detection_f1",
|
| 2131 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 2132 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 2133 |
+
"reason": null
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
"task_number": 12,
|
|
|
|
| 3416 |
"task_label": "Time-to-Next-Transition Regression",
|
| 3417 |
"series_id": "qwen3_omni_v6_lora",
|
| 3418 |
"method": "Qwen3-Omni v6 LoRA",
|
| 3419 |
+
"status": "scored",
|
| 3420 |
+
"status_label": "scored",
|
| 3421 |
+
"scored": true,
|
| 3422 |
"proxy_scored": false,
|
| 3423 |
+
"raw": 134.0687422166874,
|
| 3424 |
+
"raw_text": "134.07",
|
| 3425 |
+
"normalized_score": 0.07859666766782253,
|
| 3426 |
+
"metric_key": "time_to_transition_mae",
|
| 3427 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 3428 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 3429 |
+
"reason": null
|
| 3430 |
},
|
| 3431 |
{
|
| 3432 |
"task_number": 20,
|
metrics/task_surface_integrity.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-
|
| 4 |
"summary": {
|
| 5 |
"task_count": 12,
|
| 6 |
"expected_task_count": 12,
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:26+00:00",
|
| 4 |
"summary": {
|
| 5 |
"task_count": 12,
|
| 6 |
"expected_task_count": 12,
|
metrics/unified_task_model_radar.json
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
{
|
| 2 |
"title": "Unified 20-Task Model Radar",
|
| 3 |
"status": "pass",
|
| 4 |
-
"generated_at_utc": "2026-06-
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
-
"scored_method_task_count":
|
| 9 |
"normalization_policy": {
|
| 10 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 11 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
@@ -170,17 +170,17 @@
|
|
| 170 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 171 |
"plotted_as": "colored point overlay",
|
| 172 |
"result_record_count": 20,
|
| 173 |
-
"scored_task_count":
|
| 174 |
-
"covered_task_count":
|
| 175 |
"proxy_scored_task_count": 0,
|
| 176 |
-
"scoreless_task_count":
|
| 177 |
"unsupported_task_count": 0,
|
| 178 |
-
"not_evaluated_task_count":
|
| 179 |
"status_counts": {
|
| 180 |
-
"not_evaluated_in_verified_package":
|
| 181 |
-
"scored":
|
| 182 |
},
|
| 183 |
-
"coverage_fraction": 0.
|
| 184 |
"result_record_fraction": 1.0
|
| 185 |
},
|
| 186 |
{
|
|
@@ -1375,6 +1375,17 @@
|
|
| 1375 |
"raw_text": "0.8520",
|
| 1376 |
"status_label": "scored"
|
| 1377 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1378 |
"metadata128_simple": {
|
| 1379 |
"raw": 0.4198864140782312,
|
| 1380 |
"metric_key": "f1",
|
|
@@ -1419,17 +1430,6 @@
|
|
| 1419 |
"raw_text": "n/a",
|
| 1420 |
"status_label": "not supported"
|
| 1421 |
},
|
| 1422 |
-
"qwen3_omni_v6_lora": {
|
| 1423 |
-
"raw": null,
|
| 1424 |
-
"metric_key": "f1",
|
| 1425 |
-
"source": null,
|
| 1426 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1427 |
-
"status": "not_evaluated_in_verified_package",
|
| 1428 |
-
"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",
|
| 1429 |
-
"normalized_score": null,
|
| 1430 |
-
"raw_text": "n/a",
|
| 1431 |
-
"status_label": "not evaluated"
|
| 1432 |
-
},
|
| 1433 |
"cosmos3_super_reasoner": {
|
| 1434 |
"raw": null,
|
| 1435 |
"metric_key": "f1",
|
|
@@ -1486,6 +1486,17 @@
|
|
| 1486 |
"raw_text": "0.7153",
|
| 1487 |
"status_label": "scored"
|
| 1488 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1489 |
"metadata128_simple": {
|
| 1490 |
"raw": null,
|
| 1491 |
"metric_key": "f1",
|
|
@@ -1530,17 +1541,6 @@
|
|
| 1530 |
"raw_text": "n/a",
|
| 1531 |
"status_label": "not supported"
|
| 1532 |
},
|
| 1533 |
-
"qwen3_omni_v6_lora": {
|
| 1534 |
-
"raw": null,
|
| 1535 |
-
"metric_key": "f1",
|
| 1536 |
-
"source": null,
|
| 1537 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 1538 |
-
"status": "not_evaluated_in_verified_package",
|
| 1539 |
-
"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",
|
| 1540 |
-
"normalized_score": null,
|
| 1541 |
-
"raw_text": "n/a",
|
| 1542 |
-
"status_label": "not evaluated"
|
| 1543 |
-
},
|
| 1544 |
"cosmos3_super_reasoner": {
|
| 1545 |
"raw": null,
|
| 1546 |
"metric_key": "f1",
|
|
@@ -2374,6 +2374,17 @@
|
|
| 2374 |
"raw_text": "10.55",
|
| 2375 |
"status_label": "scored"
|
| 2376 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2377 |
"raw128_simple": {
|
| 2378 |
"raw": 52.32759475708008,
|
| 2379 |
"metric_key": "mae",
|
|
@@ -2418,17 +2429,6 @@
|
|
| 2418 |
"raw_text": "n/a",
|
| 2419 |
"status_label": "not supported"
|
| 2420 |
},
|
| 2421 |
-
"qwen3_omni_v6_lora": {
|
| 2422 |
-
"raw": null,
|
| 2423 |
-
"metric_key": "mae",
|
| 2424 |
-
"source": null,
|
| 2425 |
-
"scope": "multi_episode_128_partial_model_overlay",
|
| 2426 |
-
"status": "not_evaluated_in_verified_package",
|
| 2427 |
-
"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",
|
| 2428 |
-
"normalized_score": null,
|
| 2429 |
-
"raw_text": "n/a",
|
| 2430 |
-
"status_label": "not evaluated"
|
| 2431 |
-
},
|
| 2432 |
"cosmos3_super_reasoner": {
|
| 2433 |
"raw": null,
|
| 2434 |
"metric_key": "mae",
|
|
@@ -2492,7 +2492,7 @@
|
|
| 2492 |
"title": "Qwen3-Omni v6 LoRA",
|
| 2493 |
"status": "verified",
|
| 2494 |
"task_aligned_axes": "Qwen3",
|
| 2495 |
-
"coverage": "20 records /
|
| 2496 |
"headline": "JSON validity 0.9990; action macro-F1 0.0029",
|
| 2497 |
"source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_multiscale_cap96_v6_rank64_lr5e5_full8gpu_lora_eval_test_full/eval/metrics.json"
|
| 2498 |
},
|
|
@@ -4256,17 +4256,17 @@
|
|
| 4256 |
"task_label": "Temporal Order Verification",
|
| 4257 |
"series_id": "qwen3_omni_v6_lora",
|
| 4258 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4259 |
-
"status": "
|
| 4260 |
-
"status_label": "
|
| 4261 |
-
"scored":
|
| 4262 |
"proxy_scored": false,
|
| 4263 |
-
"raw":
|
| 4264 |
-
"raw_text": "
|
| 4265 |
-
"normalized_score":
|
| 4266 |
-
"metric_key": "
|
| 4267 |
-
"source":
|
| 4268 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4269 |
-
"reason":
|
| 4270 |
},
|
| 4271 |
{
|
| 4272 |
"task_number": 11,
|
|
@@ -4418,17 +4418,17 @@
|
|
| 4418 |
"task_label": "Multimodal Synchronization Detection",
|
| 4419 |
"series_id": "qwen3_omni_v6_lora",
|
| 4420 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4421 |
-
"status": "
|
| 4422 |
-
"status_label": "
|
| 4423 |
-
"scored":
|
| 4424 |
"proxy_scored": false,
|
| 4425 |
-
"raw":
|
| 4426 |
-
"raw_text": "
|
| 4427 |
-
"normalized_score":
|
| 4428 |
-
"metric_key": "
|
| 4429 |
-
"source":
|
| 4430 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4431 |
-
"reason":
|
| 4432 |
},
|
| 4433 |
{
|
| 4434 |
"task_number": 12,
|
|
@@ -5714,17 +5714,17 @@
|
|
| 5714 |
"task_label": "Time-to-Next-Transition Regression",
|
| 5715 |
"series_id": "qwen3_omni_v6_lora",
|
| 5716 |
"method": "Qwen3-Omni v6 LoRA",
|
| 5717 |
-
"status": "
|
| 5718 |
-
"status_label": "
|
| 5719 |
-
"scored":
|
| 5720 |
"proxy_scored": false,
|
| 5721 |
-
"raw":
|
| 5722 |
-
"raw_text": "
|
| 5723 |
-
"normalized_score":
|
| 5724 |
-
"metric_key": "
|
| 5725 |
-
"source":
|
| 5726 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 5727 |
-
"reason":
|
| 5728 |
},
|
| 5729 |
{
|
| 5730 |
"task_number": 20,
|
|
|
|
| 1 |
{
|
| 2 |
"title": "Unified 20-Task Model Radar",
|
| 3 |
"status": "pass",
|
| 4 |
+
"generated_at_utc": "2026-06-17T21:17:41+00:00",
|
| 5 |
"task_count": 20,
|
| 6 |
"method_count": 9,
|
| 7 |
"method_task_record_count": 180,
|
| 8 |
+
"scored_method_task_count": 119,
|
| 9 |
"normalization_policy": {
|
| 10 |
"higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
|
| 11 |
"lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
|
|
|
|
| 170 |
"method_detail": "Verified held-out Qwen3-Omni v6 LoRA metrics, plus task 16 and any completed private-GPU future-task probes scored from task-specific JSON.",
|
| 171 |
"plotted_as": "colored point overlay",
|
| 172 |
"result_record_count": 20,
|
| 173 |
+
"scored_task_count": 13,
|
| 174 |
+
"covered_task_count": 13,
|
| 175 |
"proxy_scored_task_count": 0,
|
| 176 |
+
"scoreless_task_count": 7,
|
| 177 |
"unsupported_task_count": 0,
|
| 178 |
+
"not_evaluated_task_count": 7,
|
| 179 |
"status_counts": {
|
| 180 |
+
"not_evaluated_in_verified_package": 7,
|
| 181 |
+
"scored": 13
|
| 182 |
},
|
| 183 |
+
"coverage_fraction": 0.65,
|
| 184 |
"result_record_fraction": 1.0
|
| 185 |
},
|
| 186 |
{
|
|
|
|
| 1375 |
"raw_text": "0.8520",
|
| 1376 |
"status_label": "scored"
|
| 1377 |
},
|
| 1378 |
+
"qwen3_omni_v6_lora": {
|
| 1379 |
+
"raw": 0.40984631701404173,
|
| 1380 |
+
"metric_key": "temporal_order_f1",
|
| 1381 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 1382 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 1383 |
+
"status": "scored",
|
| 1384 |
+
"reason": null,
|
| 1385 |
+
"normalized_score": 0.40984631701404173,
|
| 1386 |
+
"raw_text": "0.4098",
|
| 1387 |
+
"status_label": "scored"
|
| 1388 |
+
},
|
| 1389 |
"metadata128_simple": {
|
| 1390 |
"raw": 0.4198864140782312,
|
| 1391 |
"metric_key": "f1",
|
|
|
|
| 1430 |
"raw_text": "n/a",
|
| 1431 |
"status_label": "not supported"
|
| 1432 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1433 |
"cosmos3_super_reasoner": {
|
| 1434 |
"raw": null,
|
| 1435 |
"metric_key": "f1",
|
|
|
|
| 1486 |
"raw_text": "0.7153",
|
| 1487 |
"status_label": "scored"
|
| 1488 |
},
|
| 1489 |
+
"qwen3_omni_v6_lora": {
|
| 1490 |
+
"raw": 0.3344936184319576,
|
| 1491 |
+
"metric_key": "misalignment_detection_f1",
|
| 1492 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 1493 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 1494 |
+
"status": "scored",
|
| 1495 |
+
"reason": null,
|
| 1496 |
+
"normalized_score": 0.3344936184319576,
|
| 1497 |
+
"raw_text": "0.3345",
|
| 1498 |
+
"status_label": "scored"
|
| 1499 |
+
},
|
| 1500 |
"metadata128_simple": {
|
| 1501 |
"raw": null,
|
| 1502 |
"metric_key": "f1",
|
|
|
|
| 1541 |
"raw_text": "n/a",
|
| 1542 |
"status_label": "not supported"
|
| 1543 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1544 |
"cosmos3_super_reasoner": {
|
| 1545 |
"raw": null,
|
| 1546 |
"metric_key": "f1",
|
|
|
|
| 2374 |
"raw_text": "10.55",
|
| 2375 |
"status_label": "scored"
|
| 2376 |
},
|
| 2377 |
+
"qwen3_omni_v6_lora": {
|
| 2378 |
+
"raw": 134.0687422166874,
|
| 2379 |
+
"metric_key": "time_to_transition_mae",
|
| 2380 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 2381 |
+
"scope": "multi_episode_128_partial_model_overlay",
|
| 2382 |
+
"status": "scored",
|
| 2383 |
+
"reason": null,
|
| 2384 |
+
"normalized_score": 0.07859666766782253,
|
| 2385 |
+
"raw_text": "134.07",
|
| 2386 |
+
"status_label": "scored"
|
| 2387 |
+
},
|
| 2388 |
"raw128_simple": {
|
| 2389 |
"raw": 52.32759475708008,
|
| 2390 |
"metric_key": "mae",
|
|
|
|
| 2429 |
"raw_text": "n/a",
|
| 2430 |
"status_label": "not supported"
|
| 2431 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2432 |
"cosmos3_super_reasoner": {
|
| 2433 |
"raw": null,
|
| 2434 |
"metric_key": "mae",
|
|
|
|
| 2492 |
"title": "Qwen3-Omni v6 LoRA",
|
| 2493 |
"status": "verified",
|
| 2494 |
"task_aligned_axes": "Qwen3",
|
| 2495 |
+
"coverage": "20 records / 13 scored task-aligned axes",
|
| 2496 |
"headline": "JSON validity 0.9990; action macro-F1 0.0029",
|
| 2497 |
"source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_multiscale_cap96_v6_rank64_lr5e5_full8gpu_lora_eval_test_full/eval/metrics.json"
|
| 2498 |
},
|
|
|
|
| 4256 |
"task_label": "Temporal Order Verification",
|
| 4257 |
"series_id": "qwen3_omni_v6_lora",
|
| 4258 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4259 |
+
"status": "scored",
|
| 4260 |
+
"status_label": "scored",
|
| 4261 |
+
"scored": true,
|
| 4262 |
"proxy_scored": false,
|
| 4263 |
+
"raw": 0.40984631701404173,
|
| 4264 |
+
"raw_text": "0.4098",
|
| 4265 |
+
"normalized_score": 0.40984631701404173,
|
| 4266 |
+
"metric_key": "temporal_order_f1",
|
| 4267 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/temporal_order/metrics.json",
|
| 4268 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4269 |
+
"reason": null
|
| 4270 |
},
|
| 4271 |
{
|
| 4272 |
"task_number": 11,
|
|
|
|
| 4418 |
"task_label": "Multimodal Synchronization Detection",
|
| 4419 |
"series_id": "qwen3_omni_v6_lora",
|
| 4420 |
"method": "Qwen3-Omni v6 LoRA",
|
| 4421 |
+
"status": "scored",
|
| 4422 |
+
"status_label": "scored",
|
| 4423 |
+
"scored": true,
|
| 4424 |
"proxy_scored": false,
|
| 4425 |
+
"raw": 0.3344936184319576,
|
| 4426 |
+
"raw_text": "0.3345",
|
| 4427 |
+
"normalized_score": 0.3344936184319576,
|
| 4428 |
+
"metric_key": "misalignment_detection_f1",
|
| 4429 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/misalignment_detection/metrics.json",
|
| 4430 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 4431 |
+
"reason": null
|
| 4432 |
},
|
| 4433 |
{
|
| 4434 |
"task_number": 12,
|
|
|
|
| 5714 |
"task_label": "Time-to-Next-Transition Regression",
|
| 5715 |
"series_id": "qwen3_omni_v6_lora",
|
| 5716 |
"method": "Qwen3-Omni v6 LoRA",
|
| 5717 |
+
"status": "scored",
|
| 5718 |
+
"status_label": "scored",
|
| 5719 |
+
"scored": true,
|
| 5720 |
"proxy_scored": false,
|
| 5721 |
+
"raw": 134.0687422166874,
|
| 5722 |
+
"raw_text": "134.07",
|
| 5723 |
+
"normalized_score": 0.07859666766782253,
|
| 5724 |
+
"metric_key": "time_to_transition_mae",
|
| 5725 |
+
"source": "results/omni_finetune/xperience10m_qwen3_omni_v6_order_sync_time_probes_a100_20260617T132500Z/time_to_transition/metrics.json",
|
| 5726 |
"scope": "multi_episode_128_partial_model_overlay",
|
| 5727 |
+
"reason": null
|
| 5728 |
},
|
| 5729 |
{
|
| 5730 |
"task_number": 20,
|
metrics/website_integrity.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
-
"generated_at_utc": "2026-06-17T21:
|
| 4 |
"docs_root": "docs",
|
| 5 |
"site_base": "/ropedia-xperience-10m-task-suite/",
|
| 6 |
"summary": {
|
|
@@ -316,7 +316,7 @@
|
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"path": "data/episode128_task_model_radar.json",
|
| 319 |
-
"bytes":
|
| 320 |
"top_level_type": "dict"
|
| 321 |
},
|
| 322 |
{
|
|
@@ -486,12 +486,12 @@
|
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"path": "data/task_method_20_gap_audit.json",
|
| 489 |
-
"bytes":
|
| 490 |
"top_level_type": "dict"
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"path": "data/task_method_20_result_matrix.json",
|
| 494 |
-
"bytes":
|
| 495 |
"top_level_type": "dict"
|
| 496 |
},
|
| 497 |
{
|
|
@@ -526,7 +526,7 @@
|
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"path": "data/unified_task_model_radar.json",
|
| 529 |
-
"bytes":
|
| 530 |
"top_level_type": "dict"
|
| 531 |
},
|
| 532 |
{
|
|
@@ -566,7 +566,7 @@
|
|
| 566 |
{
|
| 567 |
"path": "assets/charts/episode128_task_model_radar.svg",
|
| 568 |
"exists": true,
|
| 569 |
-
"bytes":
|
| 570 |
"format": "SVG",
|
| 571 |
"has_viewbox": true
|
| 572 |
},
|
|
@@ -636,7 +636,7 @@
|
|
| 636 |
{
|
| 637 |
"path": "assets/charts/unified_task_model_radar.svg",
|
| 638 |
"exists": true,
|
| 639 |
-
"bytes":
|
| 640 |
"format": "SVG",
|
| 641 |
"has_viewbox": true
|
| 642 |
},
|
|
|
|
| 1 |
{
|
| 2 |
"status": "pass",
|
| 3 |
+
"generated_at_utc": "2026-06-17T21:25:27+00:00",
|
| 4 |
"docs_root": "docs",
|
| 5 |
"site_base": "/ropedia-xperience-10m-task-suite/",
|
| 6 |
"summary": {
|
|
|
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"path": "data/episode128_task_model_radar.json",
|
| 319 |
+
"bytes": 187309,
|
| 320 |
"top_level_type": "dict"
|
| 321 |
},
|
| 322 |
{
|
|
|
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"path": "data/task_method_20_gap_audit.json",
|
| 489 |
+
"bytes": 53574,
|
| 490 |
"top_level_type": "dict"
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"path": "data/task_method_20_result_matrix.json",
|
| 494 |
+
"bytes": 129707,
|
| 495 |
"top_level_type": "dict"
|
| 496 |
},
|
| 497 |
{
|
|
|
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"path": "data/unified_task_model_radar.json",
|
| 529 |
+
"bytes": 231161,
|
| 530 |
"top_level_type": "dict"
|
| 531 |
},
|
| 532 |
{
|
|
|
|
| 566 |
{
|
| 567 |
"path": "assets/charts/episode128_task_model_radar.svg",
|
| 568 |
"exists": true,
|
| 569 |
+
"bytes": 44378,
|
| 570 |
"format": "SVG",
|
| 571 |
"has_viewbox": true
|
| 572 |
},
|
|
|
|
| 636 |
{
|
| 637 |
"path": "assets/charts/unified_task_model_radar.svg",
|
| 638 |
"exists": true,
|
| 639 |
+
"bytes": 50394,
|
| 640 |
"format": "SVG",
|
| 641 |
"has_viewbox": true
|
| 642 |
},
|