cy0307 commited on
Commit
65977b7
·
verified ·
1 Parent(s): 3960604

Update task-suite public data mirrors

Browse files
data/artifact_index.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "title": "Ropedia Xperience-10M Task Suite Artifact Index",
3
- "generated_at_utc": "2026-06-16T06:28:50+00:00",
4
  "status": "pass",
5
- "artifact_count": 176,
6
  "missing": [],
7
  "by_kind": {
8
  "project_path": 14,
@@ -16,7 +16,7 @@
16
  "website_data": 6,
17
  "generated_figure": 5,
18
  "visualization_builder": 1,
19
- "model_result": 1,
20
  "result_interpretation": 5,
21
  "metrics_source": 27,
22
  "visual_evidence": 7,
@@ -465,7 +465,7 @@
465
  "shows": "Machine-readable source-alignment pass/fail check for repo, website, and HF surfaces.",
466
  "exists": true,
467
  "bytes": 4432,
468
- "sha256": "07798a6609929512ae6f9368dc2626b0a854b828682b8d0dfa0323698eb0f5bf"
469
  },
470
  {
471
  "id": "source_alignment_validator",
@@ -585,8 +585,8 @@
585
  "surface": "website_hf",
586
  "shows": "Stores normalized 20-axis radar values, raw task metrics, Qwen3/Cosmos overlay mappings, and branch-card caveats.",
587
  "exists": true,
588
- "bytes": 38085,
589
- "sha256": "039ecf90094b813984cdd439017d383e1712c97d2eca8e7060c9981e20e08f7b"
590
  },
591
  {
592
  "id": "unified_task_model_radar_chart",
@@ -596,8 +596,8 @@
596
  "surface": "website_hf",
597
  "shows": "Compares minimal and neural MLP baselines across all 20 tasks, with Qwen3/Cosmos task-aligned model overlays.",
598
  "exists": true,
599
- "bytes": 30823,
600
- "sha256": "c26b73cdfd53ec81b9b4958a03309d11b297a09c841ce38e7309c90dcdeb17b0"
601
  },
602
  {
603
  "id": "unified_task_model_radar_builder",
@@ -607,8 +607,8 @@
607
  "surface": "repo_hf",
608
  "shows": "Regenerates the direction-aware radar chart and machine-readable metric overlay JSON.",
609
  "exists": true,
610
- "bytes": 23778,
611
- "sha256": "feb3f2ca48bf05bda6e32678ad1e70a3c36296ae514bebe1e407cb7870009f2b"
612
  },
613
  {
614
  "id": "a100_128_metadata_task_baselines",
@@ -621,6 +621,17 @@
621
  "bytes": 46996,
622
  "sha256": "e52b42fef73ab8e3647ce28361cb8c01095e3e8a7228c86b09d484cfa15b1830"
623
  },
 
 
 
 
 
 
 
 
 
 
 
624
  {
625
  "id": "research_takeaways",
626
  "title": "Research takeaways",
@@ -717,8 +728,8 @@
717
  "surface": "repo_hf",
718
  "shows": "Catalogs public figures, charts, modality thumbnails, dimensions, hashes, roles, and source scripts.",
719
  "exists": true,
720
- "bytes": 5619,
721
- "sha256": "4e22701530f12319291f611ccb1b23b647d3905bc6b4fc13d971fc1e5a6ea89d"
722
  },
723
  {
724
  "id": "figure_index_json",
@@ -728,8 +739,8 @@
728
  "surface": "website_hf",
729
  "shows": "Machine-readable visual asset index for website and Hugging Face mirrors.",
730
  "exists": true,
731
- "bytes": 15731,
732
- "sha256": "e12260ee4d0f8a8c74700398df555100745dcecfa4f258c4fd6b68925512a1c5"
733
  },
734
  {
735
  "id": "figure_index_builder",
@@ -806,7 +817,7 @@
806
  "shows": "Machine-readable release-check summary for validators, mirrors, and public project surfaces.",
807
  "exists": true,
808
  "bytes": 8100,
809
- "sha256": "6ffa21b6c04d5ae843310ec2fa6856ed1d9f40797575a417471cec17fc447c89"
810
  },
811
  {
812
  "id": "public_surface_qa",
@@ -817,7 +828,7 @@
817
  "shows": "Keeps the repo, website, and Hugging Face cards aligned as one cohesive research project surface.",
818
  "exists": true,
819
  "bytes": 1939,
820
- "sha256": "6172c436593244cb8c1524f4a51e0d53c3a90b1126e16f8bf5c2a675e9800861"
821
  },
822
  {
823
  "id": "public_surface_qa_json",
@@ -828,7 +839,7 @@
828
  "volatile": true,
829
  "shows": "Machine-readable report for SEO/social metadata, accessible tab semantics, public links, project links, and clear project presentation.",
830
  "exists": true,
831
- "bytes": 5882,
832
  "hash_policy": "existence_and_size_only"
833
  },
834
  {
@@ -909,7 +920,7 @@
909
  "volatile": true,
910
  "shows": "Records the last live GitHub/HF URL verification after upload.",
911
  "exists": true,
912
- "bytes": 131863,
913
  "hash_policy": "existence_and_size_only"
914
  },
915
  {
@@ -953,8 +964,8 @@
953
  "surface": "repo_hf",
954
  "shows": "Generates the selective artifact catalog from local files.",
955
  "exists": true,
956
- "bytes": 48219,
957
- "sha256": "c8a612f4b1ade34cfe41abcfe5515da77c0dbe3cbcf3344b4b8a079eac1e0493"
958
  },
959
  {
960
  "id": "publication_audit",
@@ -965,7 +976,7 @@
965
  "volatile": true,
966
  "shows": "Confirms public bundles exclude raw data, caches, heavy archives, and credential text.",
967
  "exists": true,
968
- "bytes": 7881,
969
  "hash_policy": "existence_and_size_only"
970
  },
971
  {
@@ -989,7 +1000,7 @@
989
  "volatile": true,
990
  "shows": "Confirms prepared GitHub/HF Space/artifact/model mirrors share the same critical data, figure, website HTML, and validator files.",
991
  "exists": true,
992
- "bytes": 635505,
993
  "hash_policy": "existence_and_size_only"
994
  },
995
  {
 
1
  {
2
  "title": "Ropedia Xperience-10M Task Suite Artifact Index",
3
+ "generated_at_utc": "2026-06-16T07:51:12+00:00",
4
  "status": "pass",
5
+ "artifact_count": 177,
6
  "missing": [],
7
  "by_kind": {
8
  "project_path": 14,
 
16
  "website_data": 6,
17
  "generated_figure": 5,
18
  "visualization_builder": 1,
19
+ "model_result": 2,
20
  "result_interpretation": 5,
21
  "metrics_source": 27,
22
  "visual_evidence": 7,
 
465
  "shows": "Machine-readable source-alignment pass/fail check for repo, website, and HF surfaces.",
466
  "exists": true,
467
  "bytes": 4432,
468
+ "sha256": "fb8bb8e5cd7eeea8d0b5ffb22ddedb7ea36fd023d57d54b5a9c1cf0cf83a2102"
469
  },
470
  {
471
  "id": "source_alignment_validator",
 
585
  "surface": "website_hf",
586
  "shows": "Stores normalized 20-axis radar values, raw task metrics, Qwen3/Cosmos overlay mappings, and branch-card caveats.",
587
  "exists": true,
588
+ "bytes": 54075,
589
+ "sha256": "86d82d6cb9dde65aded4fbfe4d621b511f4a7e7630aef357ce270e1bc37ab50a"
590
  },
591
  {
592
  "id": "unified_task_model_radar_chart",
 
596
  "surface": "website_hf",
597
  "shows": "Compares minimal and neural MLP baselines across all 20 tasks, with Qwen3/Cosmos task-aligned model overlays.",
598
  "exists": true,
599
+ "bytes": 37507,
600
+ "sha256": "4bb600ac9ce7901a3c94ca5851f6975751b0a7b6e1d72f397565ecfb8dfba0df"
601
  },
602
  {
603
  "id": "unified_task_model_radar_builder",
 
607
  "surface": "repo_hf",
608
  "shows": "Regenerates the direction-aware radar chart and machine-readable metric overlay JSON.",
609
  "exists": true,
610
+ "bytes": 27128,
611
+ "sha256": "8c32b725318f5a3e8ba384b496e099d93f353d009e475dbc6243d229aff4ea13"
612
  },
613
  {
614
  "id": "a100_128_metadata_task_baselines",
 
621
  "bytes": 46996,
622
  "sha256": "e52b42fef73ab8e3647ce28361cb8c01095e3e8a7228c86b09d484cfa15b1830"
623
  },
624
+ {
625
+ "id": "a100_128_raw20_task_baselines",
626
+ "title": "128-episode raw-feature 20-task baselines",
627
+ "path": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/run_summary_all.json",
628
+ "kind": "model_result",
629
+ "surface": "repo_hf",
630
+ "shows": "Rerun of simple and neural baselines over 34,269 windows and staged 4430-dimensional sensor NPZ features; covers 18 of 20 task axes with documented raw-source gaps for interaction text and camera-view sync.",
631
+ "exists": true,
632
+ "bytes": 11638,
633
+ "sha256": "a2c8a63f04f9f13f97929551551505898b1c8b0b4a8964bdd0a035d5f600f4b6"
634
+ },
635
  {
636
  "id": "research_takeaways",
637
  "title": "Research takeaways",
 
728
  "surface": "repo_hf",
729
  "shows": "Catalogs public figures, charts, modality thumbnails, dimensions, hashes, roles, and source scripts.",
730
  "exists": true,
731
+ "bytes": 5614,
732
+ "sha256": "5a71c79b2c6b764a833553fa711cc0688c4d6d41edc79b1db1530a8223fdcaea"
733
  },
734
  {
735
  "id": "figure_index_json",
 
739
  "surface": "website_hf",
740
  "shows": "Machine-readable visual asset index for website and Hugging Face mirrors.",
741
  "exists": true,
742
+ "bytes": 15726,
743
+ "sha256": "587fc0e7e9e93f06341f520acf89b7ef81eeb98caaa90135881bc0d681ac2679"
744
  },
745
  {
746
  "id": "figure_index_builder",
 
817
  "shows": "Machine-readable release-check summary for validators, mirrors, and public project surfaces.",
818
  "exists": true,
819
  "bytes": 8100,
820
+ "sha256": "52c8d858159ef519fae7382a3cac2ae7edd755f22e8f3ac3b8a63a7a90dba96a"
821
  },
822
  {
823
  "id": "public_surface_qa",
 
828
  "shows": "Keeps the repo, website, and Hugging Face cards aligned as one cohesive research project surface.",
829
  "exists": true,
830
  "bytes": 1939,
831
+ "sha256": "516ca8e7b67460a6151785bfa186165f37e394ab56114bab05481c878265ffe5"
832
  },
833
  {
834
  "id": "public_surface_qa_json",
 
839
  "volatile": true,
840
  "shows": "Machine-readable report for SEO/social metadata, accessible tab semantics, public links, project links, and clear project presentation.",
841
  "exists": true,
842
+ "bytes": 5806,
843
  "hash_policy": "existence_and_size_only"
844
  },
845
  {
 
920
  "volatile": true,
921
  "shows": "Records the last live GitHub/HF URL verification after upload.",
922
  "exists": true,
923
+ "bytes": 131884,
924
  "hash_policy": "existence_and_size_only"
925
  },
926
  {
 
964
  "surface": "repo_hf",
965
  "shows": "Generates the selective artifact catalog from local files.",
966
  "exists": true,
967
+ "bytes": 48739,
968
+ "sha256": "c0229653930917bd2cd71daf763ed85f140614bdeeef1db06b712519a684e746"
969
  },
970
  {
971
  "id": "publication_audit",
 
976
  "volatile": true,
977
  "shows": "Confirms public bundles exclude raw data, caches, heavy archives, and credential text.",
978
  "exists": true,
979
+ "bytes": 7883,
980
  "hash_policy": "existence_and_size_only"
981
  },
982
  {
 
1000
  "volatile": true,
1001
  "shows": "Confirms prepared GitHub/HF Space/artifact/model mirrors share the same critical data, figure, website HTML, and validator files.",
1002
  "exists": true,
1003
+ "bytes": 727544,
1004
  "hash_policy": "existence_and_size_only"
1005
  },
1006
  {
data/figure_index.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "title": "Ropedia Xperience-10M Figure Index",
3
  "status": "pass",
4
- "generated_at_utc": "2026-06-16T06:28:50+00:00",
5
  "scope": "Public figures, diagrams, charts, and derived modality thumbnails. Raw Xperience-10M videos, annotations, RRD files, and Qwen weights are excluded.",
6
  "figure_count": 24,
7
  "figures": [
@@ -359,13 +359,13 @@
359
  "source_script": "scripts/build_unified_task_model_radar.py",
360
  "surface": "website unified task section, README, HF mirrors",
361
  "exists": true,
362
- "bytes": 30823,
363
- "sha256": "c26b73cdfd53ec81b9b4958a03309d11b297a09c841ce38e7309c90dcdeb17b0",
364
  "dimensions": {
365
  "format": "SVG",
366
  "width": 1720,
367
- "height": 1360,
368
- "view_box": "0 0 1720 1360"
369
  },
370
  "source_script_exists": true
371
  },
 
1
  {
2
  "title": "Ropedia Xperience-10M Figure Index",
3
  "status": "pass",
4
+ "generated_at_utc": "2026-06-16T07:50:07+00:00",
5
  "scope": "Public figures, diagrams, charts, and derived modality thumbnails. Raw Xperience-10M videos, annotations, RRD files, and Qwen weights are excluded.",
6
  "figure_count": 24,
7
  "figures": [
 
359
  "source_script": "scripts/build_unified_task_model_radar.py",
360
  "surface": "website unified task section, README, HF mirrors",
361
  "exists": true,
362
+ "bytes": 37499,
363
+ "sha256": "2fdcb43457b95b467fd83b9679217177d34572461649d6ed74a9831b6eb1e575",
364
  "dimensions": {
365
  "format": "SVG",
366
  "width": 1720,
367
+ "height": 1500,
368
+ "view_box": "0 0 1720 1500"
369
  },
370
  "source_script_exists": true
371
  },
data/mirror_parity.json CHANGED
The diff for this file is too large to render. See raw diff
 
data/public_surface_qa.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "title": "Ropedia Xperience-10M Public Project Surface",
3
  "status": "pass",
4
- "generated_at_utc": "2026-06-16T06:36:03+00:00",
5
  "scope": "Repo README, GitHub Pages HTML, Hugging Face Space card, artifact dataset card, and model card.",
6
  "checks": [
7
  {
@@ -18,7 +18,7 @@
18
  "website_integrity": {
19
  "exists": true,
20
  "status": "pass",
21
- "generated_at_utc": "2026-06-16T06:32:53+00:00"
22
  },
23
  "rendered_site_check": {
24
  "exists": true,
@@ -28,27 +28,27 @@
28
  "task_surface_integrity": {
29
  "exists": true,
30
  "status": "pass",
31
- "generated_at_utc": "2026-06-16T06:32:52+00:00"
32
  },
33
  "source_alignment": {
34
  "exists": true,
35
  "status": "pass",
36
- "generated_at_utc": "2026-06-16T06:32:52+00:00"
37
  },
38
  "scale_up_status": {
39
  "exists": true,
40
  "status": "pass",
41
- "generated_at_utc": "2026-06-16T06:33:02+00:00"
42
  },
43
  "publication_package": {
44
  "exists": true,
45
  "status": "pass",
46
- "generated_at_utc": "2026-06-16T06:33:43+00:00"
47
  },
48
  "mirror_parity": {
49
  "exists": true,
50
  "status": "pass",
51
- "generated_at_utc": "2026-06-16T06:33:38+00:00"
52
  }
53
  },
54
  "failures": {}
 
1
  {
2
  "title": "Ropedia Xperience-10M Public Project Surface",
3
  "status": "pass",
4
+ "generated_at_utc": "2026-06-16T07:57:01+00:00",
5
  "scope": "Repo README, GitHub Pages HTML, Hugging Face Space card, artifact dataset card, and model card.",
6
  "checks": [
7
  {
 
18
  "website_integrity": {
19
  "exists": true,
20
  "status": "pass",
21
+ "generated_at_utc": "2026-06-16T07:52:05+00:00"
22
  },
23
  "rendered_site_check": {
24
  "exists": true,
 
28
  "task_surface_integrity": {
29
  "exists": true,
30
  "status": "pass",
31
+ "generated_at_utc": "2026-06-16T07:54:01+00:00"
32
  },
33
  "source_alignment": {
34
  "exists": true,
35
  "status": "pass",
36
+ "generated_at_utc": "2026-06-16T07:54:02+00:00"
37
  },
38
  "scale_up_status": {
39
  "exists": true,
40
  "status": "pass",
41
+ "generated_at_utc": "2026-06-16T07:54:11+00:00"
42
  },
43
  "publication_package": {
44
  "exists": true,
45
  "status": "pass",
46
+ "generated_at_utc": "2026-06-16T07:54:48+00:00"
47
  },
48
  "mirror_parity": {
49
  "exists": true,
50
  "status": "pass",
51
+ "generated_at_utc": "2026-06-16T07:56:47+00:00"
52
  }
53
  },
54
  "failures": {}
data/publication_audit.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "status": "pass",
3
- "generated_at_utc": "2026-06-16T06:38:03+00:00",
4
  "checks": [
5
  {
6
  "name": "required_publication_assets_present",
@@ -193,8 +193,8 @@
193
  "github_repo": {
194
  "root": "repo",
195
  "exists": true,
196
- "file_count": 1040,
197
- "text_file_count": 855,
198
  "largest_file": {
199
  "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
200
  "bytes": 55702978
@@ -204,8 +204,8 @@
204
  "hf_space_bundle": {
205
  "root": "hf_publish/space",
206
  "exists": true,
207
- "file_count": 824,
208
- "text_file_count": 678,
209
  "largest_file": {
210
  "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
211
  "bytes": 55702978
@@ -215,8 +215,8 @@
215
  "hf_artifact_bundle": {
216
  "root": "hf_publish/artifacts",
217
  "exists": true,
218
- "file_count": 1951,
219
- "text_file_count": 857,
220
  "largest_file": {
221
  "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
222
  "bytes": 55702978
@@ -226,8 +226,8 @@
226
  "hf_model_bundle": {
227
  "root": "hf_publish/model",
228
  "exists": true,
229
- "file_count": 2374,
230
- "text_file_count": 1015,
231
  "largest_file": {
232
  "path": "pytorch_model.bin",
233
  "bytes": 93495480
 
1
  {
2
  "status": "pass",
3
+ "generated_at_utc": "2026-06-16T07:54:48+00:00",
4
  "checks": [
5
  {
6
  "name": "required_publication_assets_present",
 
193
  "github_repo": {
194
  "root": "repo",
195
  "exists": true,
196
+ "file_count": 1112,
197
+ "text_file_count": 921,
198
  "largest_file": {
199
  "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
200
  "bytes": 55702978
 
204
  "hf_space_bundle": {
205
  "root": "hf_publish/space",
206
  "exists": true,
207
+ "file_count": 895,
208
+ "text_file_count": 743,
209
  "largest_file": {
210
  "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
211
  "bytes": 55702978
 
215
  "hf_artifact_bundle": {
216
  "root": "hf_publish/artifacts",
217
  "exists": true,
218
+ "file_count": 2081,
219
+ "text_file_count": 922,
220
  "largest_file": {
221
  "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
222
  "bytes": 55702978
 
226
  "hf_model_bundle": {
227
  "root": "hf_publish/model",
228
  "exists": true,
229
+ "file_count": 2510,
230
+ "text_file_count": 1080,
231
  "largest_file": {
232
  "path": "pytorch_model.bin",
233
  "bytes": 93495480
data/quality_gates.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "title": "Ropedia Xperience-10M Release Checks",
3
  "status": "pass",
4
- "generated_at_utc": "2026-06-16T06:36:08+00:00",
5
  "rule": "A release is current when the automated reports pass and the live GitHub/Hugging Face mirrors are verified after publishing.",
6
  "automated_gates": [
7
  {
 
1
  {
2
  "title": "Ropedia Xperience-10M Release Checks",
3
  "status": "pass",
4
+ "generated_at_utc": "2026-06-16T07:50:01+00:00",
5
  "rule": "A release is current when the automated reports pass and the live GitHub/Hugging Face mirrors are verified after publishing.",
6
  "automated_gates": [
7
  {
data/scope_claims_audit.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "status": "pass",
3
- "generated_at_utc": "2026-06-16T06:36:49+00:00",
4
  "summary": {
5
  "qwen3_omni_verified_diagnostic_pilot": true,
6
  "dataset_manifest_num_episodes": 119,
 
1
  {
2
  "status": "pass",
3
+ "generated_at_utc": "2026-06-16T07:54:11+00:00",
4
  "summary": {
5
  "qwen3_omni_verified_diagnostic_pilot": true,
6
  "dataset_manifest_num_episodes": 119,
data/source_alignment_audit.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "title": "Ropedia Xperience-10M Source Alignment Note",
3
  "status": "pass",
4
- "generated_at_utc": "2026-06-16T06:36:43+00:00",
5
  "alignment_json": "docs/data/xperience10m_dataset_card_alignment.json",
6
  "alignment_summary": {
7
  "full_dataset_repo": "ropedia-ai/xperience-10m",
 
1
  {
2
  "title": "Ropedia Xperience-10M Source Alignment Note",
3
  "status": "pass",
4
+ "generated_at_utc": "2026-06-16T07:54:02+00:00",
5
  "alignment_json": "docs/data/xperience10m_dataset_card_alignment.json",
6
  "alignment_summary": {
7
  "full_dataset_repo": "ropedia-ai/xperience-10m",
data/task_surface_integrity.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "status": "pass",
3
- "generated_at_utc": "2026-06-16T06:36:43+00:00",
4
  "summary": {
5
  "task_count": 12,
6
  "expected_task_count": 12,
 
1
  {
2
  "status": "pass",
3
+ "generated_at_utc": "2026-06-16T07:54:01+00:00",
4
  "summary": {
5
  "task_count": 12,
6
  "expected_task_count": 12,
data/unified_task_model_radar.json CHANGED
@@ -1,14 +1,15 @@
1
  {
2
  "title": "Unified 20-Task Model Radar",
3
  "status": "pass",
4
- "generated_at_utc": "2026-06-16T06:32:05+00:00",
5
  "task_count": 20,
6
  "normalization_policy": {
7
  "higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
8
  "lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
9
  "raw_values": "raw metric values, metric keys, and sources are retained in this JSON; the SVG is an overview, not a replacement for the metric table",
10
  "foundation_model_overlay": "Qwen3/Cosmos points are plotted only on task-aligned axes. Missing axes mean the public result does not evaluate that task contract.",
11
- "metadata_128_overlay": "128-episode metadata baselines are plotted only where the public JSONL contains enough task labels without raw feature blocks."
 
12
  },
13
  "series": [
14
  {
@@ -55,6 +56,28 @@
55
  "covered_task_count": 6,
56
  "coverage_fraction": 0.3
57
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
  {
59
  "id": "qwen3_omni_v6_lora",
60
  "label": "Qwen3-Omni v6 LoRA",
@@ -155,6 +178,22 @@
155
  "scope": "multi_episode_128_metadata_baseline",
156
  "normalized_score": 0.004175793689174209,
157
  "raw_text": "0.0042"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
158
  }
159
  }
160
  },
@@ -215,6 +254,22 @@
215
  "scope": "multi_episode_128_metadata_baseline",
216
  "normalized_score": 7.207207207207208e-05,
217
  "raw_text": "0.0001"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
218
  }
219
  }
220
  },
@@ -283,6 +338,22 @@
283
  "scope": "multi_episode_128_metadata_baseline",
284
  "normalized_score": 0.4841733292368365,
285
  "raw_text": "0.4842"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
286
  }
287
  }
288
  },
@@ -351,6 +422,22 @@
351
  "scope": "multi_episode_128_metadata_baseline",
352
  "normalized_score": 0.004910507980164745,
353
  "raw_text": "0.0049"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
354
  }
355
  }
356
  },
@@ -379,6 +466,22 @@
379
  "scope": "single_episode_public_sample",
380
  "normalized_score": 1.0,
381
  "raw_text": "0.1079"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
382
  }
383
  }
384
  },
@@ -447,6 +550,22 @@
447
  "scope": "multi_episode_128_metadata_baseline",
448
  "normalized_score": 0.5682695682695682,
449
  "raw_text": "0.5683"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
450
  }
451
  }
452
  },
@@ -507,6 +626,22 @@
507
  "scope": "multi_episode_128_metadata_baseline",
508
  "normalized_score": 0.18662723837686876,
509
  "raw_text": "0.1866"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
510
  }
511
  }
512
  },
@@ -543,6 +678,22 @@
543
  "scope": "multi_episode_128_metadata_baseline",
544
  "normalized_score": 0.002332374220713973,
545
  "raw_text": "0.0023"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
546
  }
547
  }
548
  },
@@ -579,6 +730,22 @@
579
  "scope": "multi_episode_128_partial_model_overlay",
580
  "normalized_score": 0.022138720585222767,
581
  "raw_text": "0.0221"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
582
  }
583
  }
584
  },
@@ -607,6 +774,22 @@
607
  "scope": "single_episode_public_sample",
608
  "normalized_score": 0.0,
609
  "raw_text": "-0.0102"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
610
  }
611
  }
612
  },
@@ -643,6 +826,22 @@
643
  "scope": "multi_episode_128_metadata_baseline",
644
  "normalized_score": 0.4198864140782312,
645
  "raw_text": "0.4199"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
646
  }
647
  }
648
  },
@@ -671,6 +870,22 @@
671
  "scope": "single_episode_public_sample",
672
  "normalized_score": 0.7152682255845944,
673
  "raw_text": "0.7153"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
674
  }
675
  }
676
  },
@@ -699,6 +914,22 @@
699
  "scope": "single_episode_public_sample",
700
  "normalized_score": 0.06545454545454546,
701
  "raw_text": "0.0655"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
702
  }
703
  }
704
  },
@@ -727,6 +958,22 @@
727
  "scope": "single_episode_public_sample",
728
  "normalized_score": 0.050724637681159424,
729
  "raw_text": "0.0507"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
730
  }
731
  }
732
  },
@@ -783,6 +1030,22 @@
783
  "scope": "single_episode_public_sample",
784
  "normalized_score": 0.0,
785
  "raw_text": "0.0000"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
786
  }
787
  }
788
  },
@@ -811,6 +1074,22 @@
811
  "scope": "single_episode_public_sample",
812
  "normalized_score": 0.19718309859154928,
813
  "raw_text": "0.1972"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
814
  }
815
  }
816
  },
@@ -839,6 +1118,22 @@
839
  "scope": "single_episode_public_sample",
840
  "normalized_score": 0.9879531106266066,
841
  "raw_text": "0.0426"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
842
  }
843
  }
844
  },
@@ -895,6 +1190,22 @@
895
  "scope": "single_episode_public_sample",
896
  "normalized_score": 0.9983762814568361,
897
  "raw_text": "10.55"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
898
  }
899
  }
900
  }
@@ -916,6 +1227,22 @@
916
  "headline": "compact MLP heads over metadata/text features",
917
  "source": "results/omni_finetune/a100_128_metadata_task_baselines_20260616_v2/summary_report.json"
918
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
919
  {
920
  "id": "qwen3_omni_v6_lora",
921
  "title": "Qwen3-Omni v6 LoRA",
 
1
  {
2
  "title": "Unified 20-Task Model Radar",
3
  "status": "pass",
4
+ "generated_at_utc": "2026-06-16T07:51:12+00:00",
5
  "task_count": 20,
6
  "normalization_policy": {
7
  "higher_is_better": "bounded metrics are plotted directly on 0-1 axes after clipping to [0, 1]",
8
  "lower_is_better": "lower-error metrics are converted to best_observed_value / raw_value within the same task",
9
  "raw_values": "raw metric values, metric keys, and sources are retained in this JSON; the SVG is an overview, not a replacement for the metric table",
10
  "foundation_model_overlay": "Qwen3/Cosmos points are plotted only on task-aligned axes. Missing axes mean the public result does not evaluate that task contract.",
11
+ "metadata_128_overlay": "128-episode metadata baselines are plotted only where the public JSONL contains enough task labels without raw feature blocks.",
12
+ "raw_128_overlay": "128-episode raw-feature baselines use staged sensor NPZ features and are plotted only on task axes supported by the exported feature blocks."
13
  },
14
  "series": [
15
  {
 
56
  "covered_task_count": 6,
57
  "coverage_fraction": 0.3
58
  },
59
+ {
60
+ "id": "raw128_simple",
61
+ "label": "128ep Raw Simple",
62
+ "short_label": "128-RS",
63
+ "color": "#f59e0b",
64
+ "kind": "partial_128_episode_raw_feature_baseline",
65
+ "scope": "128 selected episodes, staged 4430-dim sensor NPZ features",
66
+ "stroke_dasharray": "8 4",
67
+ "covered_task_count": 18,
68
+ "coverage_fraction": 0.9
69
+ },
70
+ {
71
+ "id": "raw128_neural_mlp",
72
+ "label": "128ep Raw NN",
73
+ "short_label": "128-RN",
74
+ "color": "#22d3ee",
75
+ "kind": "partial_128_episode_raw_feature_baseline",
76
+ "scope": "128 selected episodes, staged 4430-dim sensor NPZ features",
77
+ "stroke_dasharray": "2 5",
78
+ "covered_task_count": 18,
79
+ "coverage_fraction": 0.9
80
+ },
81
  {
82
  "id": "qwen3_omni_v6_lora",
83
  "label": "Qwen3-Omni v6 LoRA",
 
178
  "scope": "multi_episode_128_metadata_baseline",
179
  "normalized_score": 0.004175793689174209,
180
  "raw_text": "0.0042"
181
+ },
182
+ "raw128_simple": {
183
+ "raw": 0.002915061325704321,
184
+ "metric_key": "macro_f1",
185
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/timeline_action/metrics.json",
186
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
187
+ "normalized_score": 0.002915061325704321,
188
+ "raw_text": "0.0029"
189
+ },
190
+ "raw128_neural_mlp": {
191
+ "raw": 0.0014955083181204041,
192
+ "metric_key": "macro_f1",
193
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/timeline_action/metrics.json",
194
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
195
+ "normalized_score": 0.0014955083181204041,
196
+ "raw_text": "0.0015"
197
  }
198
  }
199
  },
 
254
  "scope": "multi_episode_128_metadata_baseline",
255
  "normalized_score": 7.207207207207208e-05,
256
  "raw_text": "0.0001"
257
+ },
258
+ "raw128_simple": {
259
+ "raw": 0.0,
260
+ "metric_key": "macro_f1",
261
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/timeline_subtask/metrics.json",
262
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
263
+ "normalized_score": 0.0,
264
+ "raw_text": "0.0000"
265
+ },
266
+ "raw128_neural_mlp": {
267
+ "raw": 7.35632183908046e-05,
268
+ "metric_key": "macro_f1",
269
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/timeline_subtask/metrics.json",
270
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
271
+ "normalized_score": 7.35632183908046e-05,
272
+ "raw_text": "0.0001"
273
  }
274
  }
275
  },
 
338
  "scope": "multi_episode_128_metadata_baseline",
339
  "normalized_score": 0.4841733292368365,
340
  "raw_text": "0.4842"
341
+ },
342
+ "raw128_simple": {
343
+ "raw": 0.4203613574238283,
344
+ "metric_key": "macro_f1",
345
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/transition_detection/metrics.json",
346
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
347
+ "normalized_score": 0.4203613574238283,
348
+ "raw_text": "0.4204"
349
+ },
350
+ "raw128_neural_mlp": {
351
+ "raw": 0.4902206914147213,
352
+ "metric_key": "macro_f1",
353
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/transition_detection/metrics.json",
354
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
355
+ "normalized_score": 0.4902206914147213,
356
+ "raw_text": "0.4902"
357
  }
358
  }
359
  },
 
422
  "scope": "multi_episode_128_metadata_baseline",
423
  "normalized_score": 0.004910507980164745,
424
  "raw_text": "0.0049"
425
+ },
426
+ "raw128_simple": {
427
+ "raw": 0.003285273363482094,
428
+ "metric_key": "macro_f1",
429
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/next_action/metrics.json",
430
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
431
+ "normalized_score": 0.003285273363482094,
432
+ "raw_text": "0.0033"
433
+ },
434
+ "raw128_neural_mlp": {
435
+ "raw": 0.0018477984371755407,
436
+ "metric_key": "macro_f1",
437
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/next_action/metrics.json",
438
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
439
+ "normalized_score": 0.0018477984371755407,
440
+ "raw_text": "0.0018"
441
  }
442
  }
443
  },
 
466
  "scope": "single_episode_public_sample",
467
  "normalized_score": 1.0,
468
  "raw_text": "0.1079"
469
+ },
470
+ "raw128_simple": {
471
+ "raw": 0.2729249894618988,
472
+ "metric_key": "mae",
473
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/hand_trajectory_forecast/metrics.json",
474
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
475
+ "normalized_score": 0.39516420515180267,
476
+ "raw_text": "0.2729"
477
+ },
478
+ "raw128_neural_mlp": {
479
+ "raw": 0.18475216627120972,
480
+ "metric_key": "mae",
481
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/hand_trajectory_forecast/metrics.json",
482
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
483
+ "normalized_score": 0.5837560051580399,
484
+ "raw_text": "0.1848"
485
  }
486
  }
487
  },
 
550
  "scope": "multi_episode_128_metadata_baseline",
551
  "normalized_score": 0.5682695682695682,
552
  "raw_text": "0.5683"
553
+ },
554
+ "raw128_simple": {
555
+ "raw": 0.886990707397193,
556
+ "metric_key": "macro_f1",
557
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/contact_prediction/metrics.json",
558
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
559
+ "normalized_score": 0.886990707397193,
560
+ "raw_text": "0.8870"
561
+ },
562
+ "raw128_neural_mlp": {
563
+ "raw": 1.0,
564
+ "metric_key": "macro_f1",
565
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/contact_prediction/metrics.json",
566
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
567
+ "normalized_score": 1.0,
568
+ "raw_text": "1.000"
569
  }
570
  }
571
  },
 
626
  "scope": "multi_episode_128_metadata_baseline",
627
  "normalized_score": 0.18662723837686876,
628
  "raw_text": "0.1866"
629
+ },
630
+ "raw128_simple": {
631
+ "raw": 0.0655376369662084,
632
+ "metric_key": "micro_f1",
633
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/object_relevance/metrics.json",
634
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
635
+ "normalized_score": 0.0655376369662084,
636
+ "raw_text": "0.0655"
637
+ },
638
+ "raw128_neural_mlp": {
639
+ "raw": 0.1765890386972509,
640
+ "metric_key": "micro_f1",
641
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/object_relevance/metrics.json",
642
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
643
+ "normalized_score": 0.1765890386972509,
644
+ "raw_text": "0.1766"
645
  }
646
  }
647
  },
 
678
  "scope": "multi_episode_128_metadata_baseline",
679
  "normalized_score": 0.002332374220713973,
680
  "raw_text": "0.0023"
681
+ },
682
+ "raw128_simple": {
683
+ "raw": 0.011150892823934555,
684
+ "metric_key": "mrr",
685
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/caption_grounding/metrics.json",
686
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
687
+ "normalized_score": 0.011150892823934555,
688
+ "raw_text": "0.0112"
689
+ },
690
+ "raw128_neural_mlp": {
691
+ "raw": 0.0063402121886610985,
692
+ "metric_key": "mrr",
693
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/caption_grounding/metrics.json",
694
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
695
+ "normalized_score": 0.0063402121886610985,
696
+ "raw_text": "0.0063"
697
  }
698
  }
699
  },
 
730
  "scope": "multi_episode_128_partial_model_overlay",
731
  "normalized_score": 0.022138720585222767,
732
  "raw_text": "0.0221"
733
+ },
734
+ "raw128_simple": {
735
+ "raw": 0.003459817497059703,
736
+ "metric_key": "mrr",
737
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/cross_modal_retrieval/metrics.json",
738
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
739
+ "normalized_score": 0.003459817497059703,
740
+ "raw_text": "0.0035"
741
+ },
742
+ "raw128_neural_mlp": {
743
+ "raw": 0.002535284962505102,
744
+ "metric_key": "mrr",
745
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/cross_modal_retrieval/metrics.json",
746
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
747
+ "normalized_score": 0.002535284962505102,
748
+ "raw_text": "0.0025"
749
  }
750
  }
751
  },
 
774
  "scope": "single_episode_public_sample",
775
  "normalized_score": 0.0,
776
  "raw_text": "-0.0102"
777
+ },
778
+ "raw128_simple": {
779
+ "raw": -1.3450960391924882,
780
+ "metric_key": "r2",
781
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/modality_reconstruction/metrics.json",
782
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
783
+ "normalized_score": 0.0,
784
+ "raw_text": "-1.345"
785
+ },
786
+ "raw128_neural_mlp": {
787
+ "raw": -1.3974418160502369,
788
+ "metric_key": "r2",
789
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/modality_reconstruction/metrics.json",
790
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
791
+ "normalized_score": 0.0,
792
+ "raw_text": "-1.397"
793
  }
794
  }
795
  },
 
826
  "scope": "multi_episode_128_metadata_baseline",
827
  "normalized_score": 0.4198864140782312,
828
  "raw_text": "0.4199"
829
+ },
830
+ "raw128_simple": {
831
+ "raw": 0.49824413370686593,
832
+ "metric_key": "macro_f1",
833
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/temporal_order/metrics.json",
834
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
835
+ "normalized_score": 0.49824413370686593,
836
+ "raw_text": "0.4982"
837
+ },
838
+ "raw128_neural_mlp": {
839
+ "raw": 0.8030047098504103,
840
+ "metric_key": "macro_f1",
841
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/temporal_order/metrics.json",
842
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
843
+ "normalized_score": 0.8030047098504103,
844
+ "raw_text": "0.8030"
845
  }
846
  }
847
  },
 
870
  "scope": "single_episode_public_sample",
871
  "normalized_score": 0.7152682255845944,
872
  "raw_text": "0.7153"
873
+ },
874
+ "raw128_simple": {
875
+ "raw": 0.4958867673901769,
876
+ "metric_key": "macro_f1",
877
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/misalignment_detection/metrics.json",
878
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
879
+ "normalized_score": 0.4958867673901769,
880
+ "raw_text": "0.4959"
881
+ },
882
+ "raw128_neural_mlp": {
883
+ "raw": 0.8272709077974252,
884
+ "metric_key": "macro_f1",
885
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/misalignment_detection/metrics.json",
886
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
887
+ "normalized_score": 0.8272709077974252,
888
+ "raw_text": "0.8273"
889
  }
890
  }
891
  },
 
914
  "scope": "single_episode_public_sample",
915
  "normalized_score": 0.06545454545454546,
916
  "raw_text": "0.0655"
917
+ },
918
+ "raw128_simple": {
919
+ "raw": 0.0024280172369056294,
920
+ "metric_key": "macro_f1",
921
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/long_horizon_next_action/metrics.json",
922
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
923
+ "normalized_score": 0.0024280172369056294,
924
+ "raw_text": "0.0024"
925
+ },
926
+ "raw128_neural_mlp": {
927
+ "raw": 0.001063859887389299,
928
+ "metric_key": "macro_f1",
929
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/long_horizon_next_action/metrics.json",
930
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
931
+ "normalized_score": 0.001063859887389299,
932
+ "raw_text": "0.0011"
933
  }
934
  }
935
  },
 
958
  "scope": "single_episode_public_sample",
959
  "normalized_score": 0.050724637681159424,
960
  "raw_text": "0.0507"
961
+ },
962
+ "raw128_simple": {
963
+ "raw": 0.0,
964
+ "metric_key": "macro_f1",
965
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/next_subtask_forecast/metrics.json",
966
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
967
+ "normalized_score": 0.0,
968
+ "raw_text": "0.0000"
969
+ },
970
+ "raw128_neural_mlp": {
971
+ "raw": 0.0,
972
+ "metric_key": "macro_f1",
973
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/next_subtask_forecast/metrics.json",
974
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
975
+ "normalized_score": 0.0,
976
+ "raw_text": "0.0000"
977
  }
978
  }
979
  },
 
1030
  "scope": "single_episode_public_sample",
1031
  "normalized_score": 0.0,
1032
  "raw_text": "0.0000"
1033
+ },
1034
+ "raw128_simple": {
1035
+ "raw": 0.0,
1036
+ "metric_key": "macro_f1",
1037
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/action_object_relation/metrics.json",
1038
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
1039
+ "normalized_score": 0.0,
1040
+ "raw_text": "0.0000"
1041
+ },
1042
+ "raw128_neural_mlp": {
1043
+ "raw": 0.0,
1044
+ "metric_key": "macro_f1",
1045
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/action_object_relation/metrics.json",
1046
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
1047
+ "normalized_score": 0.0,
1048
+ "raw_text": "0.0000"
1049
  }
1050
  }
1051
  },
 
1074
  "scope": "single_episode_public_sample",
1075
  "normalized_score": 0.19718309859154928,
1076
  "raw_text": "0.1972"
1077
+ },
1078
+ "raw128_simple": {
1079
+ "raw": 0.06469493412657774,
1080
+ "metric_key": "micro_f1",
1081
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/object_set_forecast/metrics.json",
1082
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
1083
+ "normalized_score": 0.06469493412657774,
1084
+ "raw_text": "0.0647"
1085
+ },
1086
+ "raw128_neural_mlp": {
1087
+ "raw": 0.17523098630012288,
1088
+ "metric_key": "micro_f1",
1089
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/object_set_forecast/metrics.json",
1090
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
1091
+ "normalized_score": 0.17523098630012288,
1092
+ "raw_text": "0.1752"
1093
  }
1094
  }
1095
  },
 
1118
  "scope": "single_episode_public_sample",
1119
  "normalized_score": 0.9879531106266066,
1120
  "raw_text": "0.0426"
1121
+ },
1122
+ "raw128_simple": {
1123
+ "raw": 0.22941437363624573,
1124
+ "metric_key": "mae",
1125
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/imu_to_hand_pose/metrics.json",
1126
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
1127
+ "normalized_score": 0.1832902066792771,
1128
+ "raw_text": "0.2294"
1129
+ },
1130
+ "raw128_neural_mlp": {
1131
+ "raw": 0.252998411655426,
1132
+ "metric_key": "mae",
1133
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/imu_to_hand_pose/metrics.json",
1134
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
1135
+ "normalized_score": 0.1662042369509182,
1136
+ "raw_text": "0.2530"
1137
  }
1138
  }
1139
  },
 
1190
  "scope": "single_episode_public_sample",
1191
  "normalized_score": 0.9983762814568361,
1192
  "raw_text": "10.55"
1193
+ },
1194
+ "raw128_simple": {
1195
+ "raw": 52.32759094238281,
1196
+ "metric_key": "mae",
1197
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/simple_raw128/time_to_transition/metrics.json",
1198
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
1199
+ "normalized_score": 0.20137285487211684,
1200
+ "raw_text": "52.33"
1201
+ },
1202
+ "raw128_neural_mlp": {
1203
+ "raw": 42.374061584472656,
1204
+ "metric_key": "mae",
1205
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/neural_mlp_raw128/time_to_transition/metrics.json",
1206
+ "scope": "multi_episode_128_raw_sensor_feature_baseline",
1207
+ "normalized_score": 0.24867468405504953,
1208
+ "raw_text": "42.37"
1209
  }
1210
  }
1211
  }
 
1227
  "headline": "compact MLP heads over metadata/text features",
1228
  "source": "results/omni_finetune/a100_128_metadata_task_baselines_20260616_v2/summary_report.json"
1229
  },
1230
+ {
1231
+ "id": "raw128_simple",
1232
+ "title": "128ep Raw Simple",
1233
+ "status": "a100_raw20_pass_with_documented_gaps",
1234
+ "coverage": "18/20 raw-feature-supported axes",
1235
+ "headline": "34,269 windows; centroid/ridge heads over 4430-dim sensor feature blocks",
1236
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/run_summary_all.json"
1237
+ },
1238
+ {
1239
+ "id": "raw128_neural_mlp",
1240
+ "title": "128ep Raw NN",
1241
+ "status": "a100_raw20_pass_with_documented_gaps",
1242
+ "coverage": "18/20 raw-feature-supported axes",
1243
+ "headline": "MLP heads ran on staged feature shards; raw interaction text and paired camera-view features are absent",
1244
+ "source": "results/omni_finetune/a100_128_raw20_task_baselines_20260616T073954Z/run_summary_all.json"
1245
+ },
1246
  {
1247
  "id": "qwen3_omni_v6_lora",
1248
  "title": "Qwen3-Omni v6 LoRA",
data/website_integrity.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "status": "pass",
3
- "generated_at_utc": "2026-06-16T06:36:48+00:00",
4
  "docs_root": "docs",
5
  "site_base": "/ropedia-xperience-10m-task-suite/",
6
  "summary": {
@@ -301,7 +301,7 @@
301
  },
302
  {
303
  "path": "data/artifact_index.json",
304
- "bytes": 96011,
305
  "top_level_type": "dict"
306
  },
307
  {
@@ -336,7 +336,7 @@
336
  },
337
  {
338
  "path": "data/live_publication_status.json",
339
- "bytes": 131863,
340
  "top_level_type": "dict"
341
  },
342
  {
@@ -491,7 +491,7 @@
491
  },
492
  {
493
  "path": "data/unified_task_model_radar.json",
494
- "bytes": 38085,
495
  "top_level_type": "dict"
496
  },
497
  {
@@ -587,7 +587,7 @@
587
  {
588
  "path": "assets/charts/unified_task_model_radar.svg",
589
  "exists": true,
590
- "bytes": 30823,
591
  "format": "SVG",
592
  "has_viewbox": true
593
  },
 
1
  {
2
  "status": "pass",
3
+ "generated_at_utc": "2026-06-16T07:52:05+00:00",
4
  "docs_root": "docs",
5
  "site_base": "/ropedia-xperience-10m-task-suite/",
6
  "summary": {
 
301
  },
302
  {
303
  "path": "data/artifact_index.json",
304
+ "bytes": 96646,
305
  "top_level_type": "dict"
306
  },
307
  {
 
336
  },
337
  {
338
  "path": "data/live_publication_status.json",
339
+ "bytes": 131884,
340
  "top_level_type": "dict"
341
  },
342
  {
 
491
  },
492
  {
493
  "path": "data/unified_task_model_radar.json",
494
+ "bytes": 54075,
495
  "top_level_type": "dict"
496
  },
497
  {
 
587
  {
588
  "path": "assets/charts/unified_task_model_radar.svg",
589
  "exists": true,
590
+ "bytes": 37507,
591
  "format": "SVG",
592
  "has_viewbox": true
593
  },