File size: 68,411 Bytes
4b0d658 c325020 4b0d658 540e67a 4b0d658 1a4adbb 79ed47f 4b0d658 7cc9dbe b7a466b 7cc9dbe b7a466b 7cc9dbe f7f39ba c325020 f7f39ba b7a466b f7f39ba c325020 f7f39ba b7a466b 46f443b b7a466b fc9e8cf b7a466b 4602161 f7f39ba 31e3087 fe4bbfa c96262b 16a39bb c96262b 16a39bb c96262b 16a39bb c96262b 16a39bb c96262b 16a39bb c96262b 16a39bb c96262b 01f57c3 2bd560e 05689bb 2bd560e 91b502e a07660e 7c58b77 b5c6bbb 98cf463 c433b73 98cf463 627e5d7 0995310 d96f266 bfcf156 4b0d658 cf07180 4b0d658 b7a466b 4b0d658 c325020 4b0d658 b7a466b 4b0d658 4bd6e11 c325020 4bd6e11 b7a466b 4bd6e11 94a5118 b7a466b 94a5118 b7a466b cca436c 29331c9 c325020 29331c9 b7a466b 29331c9 c325020 29331c9 b7a466b 29331c9 b7a466b 29331c9 b871266 b7a466b b871266 b7bdcde cca436c b7a466b cca436c b7a466b cca436c b7a466b 94a5118 d9be7c0 965d0da d9be7c0 965d0da d9be7c0 2ebe45d 47429ce ac3e830 d8565bc 05689bb d8565bc ac3e830 f52ad36 ac3e830 f52ad36 2ebe45d 13d3eec f52ad36 13d3eec f52ad36 13d3eec 5331178 77e332b 2ebe45d 47429ce 2ebe45d d8565bc 47429ce d8565bc 47429ce d8565bc 2ebe45d 47429ce 2ebe45d 13d3eec f52ad36 13d3eec 5331178 13d3eec 05689bb 13d3eec c0ec867 05689bb c0ec867 948bb27 09b6435 f135706 09b6435 f135706 09b6435 d735235 b7a466b d735235 b7a466b d735235 b7a466b d735235 ca4ac1c 45c1706 ca4ac1c 6460b80 ca4ac1c 45c1706 ca4ac1c 45c1706 ca4ac1c 6460b80 ca4ac1c 9d58132 b7a466b 9d58132 b7a466b 9d58132 b7a466b 9d58132 2c5b88c b7a466b 2c5b88c b7a466b 2c5b88c b7a466b 2c5b88c 518399e 08a4bf0 540e67a 08a4bf0 b7a466b 08a4bf0 540e67a 08a4bf0 b7a466b 08a4bf0 c4212da 540e67a c4212da b7a466b c4212da a49986a 22907ac a49986a c4212da 540e67a c4212da 04c0bde c4212da 540e67a c4212da 04c0bde c4212da 4173e02 d9be7c0 4173e02 149cadc 4173e02 b7a466b 4173e02 a6472b6 b7a466b a6472b6 11fb2f4 b7a466b 11fb2f4 7faed79 b7a466b 7faed79 b7a466b 7faed79 4b0d658 c325020 4b0d658 b7a466b 4b0d658 6a1869c 4b0d658 cf07180 4b0d658 04c0bde 4b0d658 f590d7e cf07180 6a1869c f590d7e cf07180 f590d7e 45c1706 f590d7e 0f9a8e2 b7a466b 0f9a8e2 7977885 b7a466b 7977885 4b0d658 b7a466b 4b0d658 d9be7c0 4b0d658 b7a466b 4b0d658 b7a466b 4b0d658 a8124a8 4b0d658 b7a466b 4b0d658 b7a466b 4b0d658 6460b80 4b0d658 6460b80 4b0d658 b7a466b 4b0d658 c614c4e 965d0da c614c4e 965d0da c614c4e 965d0da c614c4e 965d0da c614c4e 965d0da c614c4e 965d0da c614c4e 965d0da c614c4e 965d0da c614c4e 4b0d658 b7a466b 4b0d658 d9be7c0 4b0d658 b7a466b 4b0d658 3e04138 b7a466b 3e04138 b7a466b 3e04138 4b0d658 b7a466b 4b0d658 b7a466b 4b0d658 476e8e8 4b0d658 2bd560e 4b0d658 627e5d7 4b0d658 cfd29be 4b0d658 b7a466b 4b0d658 91b502e 627e5d7 91b502e a07660e 6460b80 a07660e 627e5d7 4b0d658 b7a466b 4b0d658 b7a466b 4b0d658 627e5d7 1a4adbb 79ed47f 4b0d658 a8124a8 627e5d7 1a4adbb 79ed47f a8124a8 4b0d658 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 | #!/usr/bin/env python3
"""Build a compact source-of-truth artifact index for the research project.
The index is intentionally selective. It lists the files behind the public
project readouts, not every prediction array or checkpoint in the repository.
"""
from __future__ import annotations
import hashlib
import json
from datetime import datetime, timezone
from pathlib import Path
ROOT = Path(__file__).resolve().parents[1]
OUTPUT = ROOT / "docs/data/artifact_index.json"
QWEN3_FUTURE_TASK_PROBE_RUN_ID = "xperience10m_qwen3_omni_v6_future_task_probes_a100_20260616T143608Z"
COSMOS3_SUPER_INTERACTION_TEXT_TASK_PROBE_RUN_ID = (
"xperience10m_cosmos3_super_interaction_text_task15_textonly_v1_20260620T1558Z"
)
ARTIFACTS = [
{
"id": "project_brief",
"title": "Project brief",
"path": "PROJECT_BRIEF.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Gives first-pass readers a concise project shape before the detailed artifact trail.",
},
{
"id": "project_brief_json",
"title": "Project brief JSON",
"path": "docs/data/project_brief.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Machine-readable first-reader project brief for the website and Hugging Face mirrors.",
},
{
"id": "project_status",
"title": "Project status",
"path": "PROJECT_STATUS.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Gives a compact current-state table for first-pass readers.",
},
{
"id": "project_status_json",
"title": "Project status JSON",
"path": "docs/data/project_status.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Machine-readable copy of the current project status for website and HF mirrors.",
},
{
"id": "glossary",
"title": "Glossary",
"path": "GLOSSARY.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Defines terminology that can be confused across data scope, task metrics, model branches, and public mirrors.",
},
{
"id": "glossary_json",
"title": "Glossary JSON",
"path": "docs/data/glossary.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Machine-readable terminology layer for the website, artifact dataset, model mirror, and public QA checks.",
},
{
"id": "research_roadmap",
"title": "Research roadmap",
"path": "RESEARCH_ROADMAP.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Defines the path from public-sample task development to multi-episode held-out evaluation and larger omni-model extensions.",
},
{
"id": "research_roadmap_json",
"title": "Research roadmap JSON",
"path": "docs/data/research_roadmap.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Machine-readable research roadmap for the website and Hugging Face mirrors.",
},
{
"id": "foundation_model_plan",
"title": "Foundation model plan",
"path": "FOUNDATION_MODEL_PLAN.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Defines the post-data-gate backbone choices: Qwen3-Omni first, Cosmos 3 for world modeling, and VLA/policy models after action-target conversion.",
},
{
"id": "foundation_model_plan_json",
"title": "Foundation model plan JSON",
"path": "docs/data/foundation_model_plan.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Machine-readable foundation-model selection matrix with source links, entry conditions, and evaluation additions.",
},
{
"id": "three_foundation_pipelines",
"title": "Three foundation pipeline tracks",
"path": "THREE_FOUNDATION_PIPELINES.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Frames spatial intelligence, human-video world modeling, and vision-language-action as three pipeline tracks with explicit inputs, outputs, maturity, and next evidence gates.",
},
{
"id": "three_foundation_pipelines_json",
"title": "Three foundation pipeline tracks JSON",
"path": "docs/data/three_foundation_pipelines.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Machine-readable pipeline-track contract for the website and Hugging Face mirrors.",
},
{
"id": "spatial_intelligence_slide_diagram",
"title": "Spatial intelligence slide diagram",
"path": "docs/assets/foundation-pipelines/spatial-intelligence-pipeline.png",
"kind": "visual_asset",
"surface": "website_hf",
"shows": "High-resolution slide diagram for the spatial intelligence model training pipeline direction.",
},
{
"id": "human_video_world_model_slide_diagram",
"title": "Human-video world model slide diagram",
"path": "docs/assets/foundation-pipelines/human-video-world-model-pipeline.png",
"kind": "visual_asset",
"surface": "website_hf",
"shows": "High-resolution slide diagram for the human-video world-model training pipeline direction.",
},
{
"id": "vision_language_action_slide_diagram",
"title": "Vision-language-action slide diagram",
"path": "docs/assets/foundation-pipelines/vision-language-action-pipeline.png",
"kind": "visual_asset",
"surface": "website_hf",
"shows": "High-resolution slide diagram for the vision-language-action training pipeline direction.",
},
{
"id": "spatial_intelligence_source_slide",
"title": "Spatial intelligence source slide",
"path": "docs/assets/foundation-pipelines/source-slides/spatial-intelligence-slide.png",
"kind": "visual_asset_source",
"surface": "repo_hf",
"shows": "Clean source slide PNG supplied for the spatial intelligence public direction figure.",
},
{
"id": "human_video_world_model_source_slide",
"title": "Human-video world model source slide",
"path": "docs/assets/foundation-pipelines/source-slides/human-video-world-model-slide.png",
"kind": "visual_asset_source",
"surface": "repo_hf",
"shows": "Clean source slide PNG supplied for the human-video world-model public direction figure.",
},
{
"id": "vision_language_action_source_slide",
"title": "Vision-language-action source slide",
"path": "docs/assets/foundation-pipelines/source-slides/vision-language-action-slide.png",
"kind": "visual_asset_source",
"surface": "repo_hf",
"shows": "Clean source slide PNG supplied for the vision-language-action public direction figure.",
},
{
"id": "omni_model_extension_contract",
"title": "Omni model extension contract",
"path": "OMNI_MODEL_EXTENSION_CONTRACT.md",
"kind": "scaleup_contract",
"surface": "repo_hf",
"shows": "Defines the shared manifest, episode split, held-out evaluation, packaging, and public-safety rules for Qwen3-Omni, Cosmos3, and VLA/policy model tracks.",
},
{
"id": "omni_backbone_registry_configs",
"title": "Omni backbone registry configs",
"path": "configs/omni_backbones",
"kind": "scaleup_contract",
"surface": "repo_hf",
"shows": "Stores the implemented Qwen3-Omni LoRA contract and planned Cosmos-style world-model and VLA/policy branch contracts.",
},
{
"id": "omni_backbone_registry_validator",
"title": "Omni backbone registry validator",
"path": "scripts/omni/backbone_registry.py",
"kind": "scaleup_contract",
"surface": "repo_hf",
"shows": "Validates backbone ids, split defaults, leakage guards, required metrics, required files, and forbidden public package categories.",
},
{
"id": "omni_model_neutral_window_index_exporter",
"title": "Model-neutral window index exporter",
"path": "scripts/omni/export_model_neutral_window_index.py",
"kind": "scaleup_contract",
"surface": "repo_hf",
"shows": "Converts Qwen JSONL records into a model-neutral window index with Qwen, Cosmos-style, and policy/VLA adapter views.",
},
{
"id": "omni_backbone_scaffolder",
"title": "Omni backbone scaffolder",
"path": "scripts/omni/scaffold_omni_backbone.py",
"kind": "scaleup_contract",
"surface": "repo_hf",
"shows": "Creates a validated planned-backbone config from an existing contract template so new model families inherit the shared rules.",
},
{
"id": "omni_backbone_packaging_smoke",
"title": "Omni backbone packaging smoke test",
"path": "scripts/omni/smoke_test_backbone_packaging.py",
"kind": "scaleup_contract",
"surface": "repo_hf",
"shows": "Builds synthetic verified packages for every configured backbone and audits them against the public-safe package contract.",
},
{
"id": "qwen3_omni_error_analysis_script",
"title": "Qwen3-Omni held-out error-analysis script",
"path": "scripts/omni/analyze_qwen3_omni_errors.py",
"kind": "scaleup_contract",
"surface": "repo_hf",
"shows": "Computes public-safe held-out error-analysis tables by episode, action family, train-seen status, required-modality state, and object category.",
},
{
"id": "multi_episode_128_baseline_script",
"title": "128-episode aligned baseline runner",
"path": "scripts/omni/run_128_task_baselines.py",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Runs simple metadata and neural MLP baselines on the same selected 96/16/16 episode split used by the Qwen3-Omni diagnostic pilot.",
},
{
"id": "task_suite_enhancement_128",
"title": "128-episode task-suite enhancement pack",
"path": "TASK_SUITE_ENHANCEMENT_128.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Records the no-new-episode dense-window, hierarchical-target, bottleneck, and experiment-backlog plan for pushing the current 128-episode suite harder without overwriting prior results.",
},
{
"id": "task_suite_enhancement_128_json",
"title": "128-episode task-suite enhancement JSON",
"path": "docs/data/task_suite_enhancement_128.json",
"kind": "scaleup_status",
"surface": "website_hf",
"shows": "Machine-readable enhancement pack for the website and Hugging Face mirrors.",
},
{
"id": "task_suite_enhancement_128_result",
"title": "128-episode task-suite enhancement result package",
"path": "results/omni_finetune/task_suite_enhancement_128_v1_20260608/enhancement_plan.json",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Versioned result directory with dense-window estimates, hierarchical target contract, task bottlenecks, Qwen action-family error summary, and experiment cards.",
},
{
"id": "task_suite_enhancement_128_builder",
"title": "128-episode task-suite enhancement builder",
"path": "scripts/omni/build_task_suite_enhancement_128.py",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Regenerates the enhancement pack from committed 128-episode windows, baseline summaries, verified Qwen predictions, and Cosmos reference metrics.",
},
{
"id": "xperience10m_128_episode_feature_index",
"title": "Xperience-10M 128-episode source and feature index",
"path": "XPERIENCE10M_128_EPISODE_FEATURE_INDEX.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Links each selected official session UUID and episode id to the gated Xperience-10M source tree plus the public-safe processed feature artifacts derived from the selected split.",
},
{
"id": "xperience10m_128_episode_feature_index_json",
"title": "Xperience-10M 128-episode source and feature index JSON",
"path": "docs/data/xperience10m_128_episode_feature_index.json",
"kind": "scaleup_status",
"surface": "website_hf",
"shows": "Machine-readable 128-episode source-to-feature map for GitHub Pages, HF Space, artifact dataset, and baseline-model mirrors.",
},
{
"id": "xperience10m_128_episode_feature_index_builder",
"title": "128-episode source and feature index builder",
"path": "scripts/omni/build_128_episode_feature_index.py",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Regenerates the public-safe source/feature index from the selected episode manifest, dense multiscale export, metadata matrices, and raw20 baseline summary.",
},
{
"id": "xperience10m_128_dense_multiscale_windows",
"title": "128-episode dense multiscale public-safe windows",
"path": "results/omni_finetune/xperience10m_128ep_dense_multiscale_hierarchical_v1_20260608/dense_multiscale_windows.jsonl",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Public-safe dense/medium/long window rows over the exportable selected episodes, linked back to official source episode ids without redistributing raw gated files.",
},
{
"id": "xperience10m_128_metadata_matrix_v2",
"title": "128-episode metadata feature matrix v2",
"path": "results/omni_finetune/a100_128_metadata_task_baselines_20260616_v2/metadata_feature_matrix.npz",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Public-safe 34,269 x 394 metadata/text feature matrix used by the aligned 128-episode metadata baseline layer.",
},
{
"id": "qwen3_full_parameter_gates",
"title": "Qwen3-Omni full-parameter feasibility gates",
"path": "results/omni_finetune/QWEN3_FULL_PARAMETER_GATES_20260609.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Summarizes the 2026-06-09 full-parameter FSDP feasibility gates: 1/8/32/64-step guarded runs passed, the 128-step opportunistic pilot was preempted for Qwen v5 handoff, and no full checkpoints or weights are published.",
},
{
"id": "qwen3_full_parameter_gates_json",
"title": "Qwen3-Omni full-parameter feasibility gates JSON",
"path": "docs/data/qwen3_full_parameter_gates.json",
"kind": "scaleup_status",
"surface": "website_hf",
"shows": "Machine-readable summary of full-parameter feasibility evidence and publication policy for website and Hugging Face mirrors.",
},
{
"id": "qwen3_v5_v6_comparison",
"title": "Qwen3-Omni v5/v6 comparison",
"path": "results/omni_finetune/QWEN3_V5_V6_COMPARISON_20260614.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Reader-facing comparison of the verified Qwen3 v5 release row and the latest verified v6 row, including metric deltas and release-tag policy.",
},
{
"id": "qwen3_v5_v6_comparison_json",
"title": "Qwen3-Omni v5/v6 comparison JSON",
"path": "docs/data/qwen3_v5_v6_comparison.json",
"kind": "scaleup_status",
"surface": "website_hf",
"shows": "Machine-readable v5/v6 metric deltas and publication recommendation for website and Hugging Face mirrors.",
},
{
"id": "qwen3_full_parameter_gates_builder",
"title": "Qwen3-Omni full-parameter gate summary builder",
"path": "scripts/omni/build_qwen3_full_parameter_gate_summary.py",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Regenerates the full-parameter feasibility-gate Markdown and JSON summaries from the run-local evidence files.",
},
{
"id": "qwen3_full_parameter_post_verified_deferrer",
"title": "Qwen3-Omni post-verified full-parameter deferrer",
"path": "scripts/omni/defer_qwen3_fullparam_after_verified_qwen.sh",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Waits for a verified Qwen held-out package, then launches a bounded 128-step full-parameter feasibility pilot on the same multiscale v5 dataset with no checkpoints or weights saved.",
},
{
"id": "qwen3_lora_hf_package_builder",
"title": "Qwen3 LoRA HF package builder",
"path": "scripts/omni/prepare_qwen3_lora_hf_package.py",
"kind": "publication_workflow",
"surface": "repo_hf",
"shows": "Builds the upload-ready Hugging Face adapter folder from a verified Qwen3 LoRA result summary and adapter directory.",
},
{
"id": "qwen3_private_gpu_repro_smoke",
"title": "Qwen3 private staged-GPU reproduction smoke",
"path": "scripts/omni/run_private_gpu_qwen3_v6_repro_smoke.sh",
"kind": "reproducibility",
"surface": "repo_hf",
"shows": "Runs the owner-side Qwen3-Omni v6 one-sample reproduction smoke from a private staged model, adapter, JSONL, and exported media cache.",
},
{
"id": "qwen3_video_feature_compat_patch",
"title": "Qwen3 video-feature compatibility patch checker",
"path": "scripts/omni/patch_qwen3_omni_video_features.py",
"kind": "reproducibility",
"surface": "repo_hf",
"shows": "Checks and narrowly repairs the installed Qwen3-Omni video-feature branch so private staged-GPU reproduction uses the verified source-compatible behavior.",
},
{
"id": "additional_development_directions",
"title": "Additional development directions",
"path": "ADDITIONAL_DEVELOPMENT_DIRECTIONS.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Records concrete non-backbone Xperience-10M development tracks: taxonomy, benchmark protocol, representation learning, skill graphs, affordances, 3D/4D memory, QA, and policy transfer.",
},
{
"id": "additional_development_directions_json",
"title": "Additional development directions JSON",
"path": "docs/data/additional_development_directions.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Machine-readable additional development directions for the website and Hugging Face mirrors.",
},
{
"id": "xperience_embodied_foundation_pretraining",
"title": "Xperience Embodied Foundation Model pretraining goal",
"path": "XPERIENCE_EMBODIED_FOUNDATION_MODEL_PRETRAINING.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Describes the future full-corpus Xperience-native pretraining goal, target modules, objectives, staged scale-up, hardware ranges, and evaluation protocol.",
},
{
"id": "evidence_contract",
"title": "Evidence contract",
"path": "EVIDENCE_CONTRACT.md",
"kind": "project_scope",
"surface": "repo",
"shows": "Defines the implemented scope, setup-stage items, and multi-episode prerequisites.",
},
{
"id": "project_packet",
"title": "Project packet",
"path": "docs/data/project_packet.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Gives a short project path with scope status and public surfaces.",
},
{
"id": "artifact_guide",
"title": "Artifact guide",
"path": "ARTIFACT_GUIDE.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Gives the human-readable map from project scope to data, tasks, platform mirrors, and scale-up status.",
},
{
"id": "official_dataset_card_alignment",
"title": "Official Xperience-10M dataset-card alignment",
"path": "XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
"kind": "source_alignment",
"surface": "repo_hf",
"shows": "Aligns public dataset wording with the official gated Xperience-10M card, public sample card, HF API metadata, and current project coverage.",
},
{
"id": "official_dataset_card_alignment_json",
"title": "Official Xperience-10M dataset-card alignment JSON",
"path": "docs/data/xperience10m_dataset_card_alignment.json",
"kind": "source_alignment",
"surface": "website_hf",
"shows": "Machine-readable upstream dataset-card, sample-card, and HF API alignment facts for website and HF mirrors.",
},
{
"id": "source_alignment",
"title": "Source alignment",
"path": "SOURCE_ALIGNMENT_AUDIT.md",
"kind": "source_alignment",
"surface": "repo_hf",
"shows": "Summarizes the pass/fail check for full-dataset facts, sample-card facts, API-listing notes, and project coverage.",
},
{
"id": "source_alignment_json",
"title": "Source alignment JSON",
"path": "docs/data/source_alignment_audit.json",
"kind": "source_alignment",
"surface": "website_hf",
"shows": "Machine-readable source-alignment pass/fail check for repo, website, and HF surfaces.",
},
{
"id": "source_alignment_validator",
"title": "Source alignment validator",
"path": "scripts/validate_source_alignment.py",
"kind": "source_alignment",
"surface": "repo_hf",
"shows": "Regenerates the source-alignment report from committed facts and public card text.",
},
{
"id": "hf_publisher",
"title": "Hugging Face publisher",
"path": "scripts/publish_hf_bundles.py",
"kind": "publication_workflow",
"surface": "repo_hf",
"shows": "Publishes prepared Space, artifact dataset, and model bundles, including an explicit model-binary upload batch.",
},
{
"id": "github_package_dockerfile",
"title": "GitHub package Dockerfile",
"path": "Dockerfile",
"kind": "publication_workflow",
"surface": "repo",
"shows": "Builds the static-dashboard container package for GitHub Container Registry.",
},
{
"id": "github_package_workflow",
"title": "GitHub package workflow",
"path": ".github/workflows/publish-ghcr.yml",
"kind": "publication_workflow",
"surface": "repo",
"shows": "Publishes the static-dashboard image to GitHub Container Registry on main or manual dispatch.",
},
{
"id": "evaluation_protocol",
"title": "Evaluation protocol",
"path": "EVALUATION_PROTOCOL.md",
"kind": "evaluation_protocol",
"surface": "repo_hf",
"shows": "Defines the window unit, chronological split, task metrics, leakage controls, and current limitations.",
},
{
"id": "evaluation_protocol_json",
"title": "Evaluation protocol JSON",
"path": "docs/data/evaluation_protocol.json",
"kind": "evaluation_protocol",
"surface": "website_hf",
"shows": "Machine-readable protocol generated from committed task metrics for website and HF mirrors.",
},
{
"id": "evaluation_protocol_builder",
"title": "Evaluation protocol builder",
"path": "scripts/build_evaluation_protocol.py",
"kind": "evaluation_protocol",
"surface": "repo_hf",
"shows": "Regenerates the protocol from committed summary metrics and task artifacts.",
},
{
"id": "task_suite_20",
"title": "Unified 20-task suite",
"path": "TASK_SUITE_20.md",
"kind": "evaluation_protocol",
"surface": "repo_hf",
"shows": "Reader-facing table for the single unified public-sample task suite under the same window, split, feature, and baseline contract.",
},
{
"id": "task_suite_20_json",
"title": "Unified 20-task suite JSON",
"path": "docs/data/task_suite_20.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Machine-readable unified 20-task index for the website, Hugging Face mirrors, and live verification.",
},
{
"id": "task_suite_20_builder",
"title": "Unified 20-task suite builder",
"path": "scripts/build_unified_task_suite.py",
"kind": "evaluation_protocol",
"surface": "repo_hf",
"shows": "Regenerates the unified 20-task JSON and Markdown from the public-sample metrics plus the historical provenance result bundle.",
},
{
"id": "unified_task_model_radar_json",
"title": "Unified 20-task model radar JSON",
"path": "docs/data/unified_task_model_radar.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Stores normalized 20-axis radar values, raw task metrics, grouped chart-design metadata, Qwen3-Omni/Cosmos3 source mappings, method-card caveats, proxy flags, and source artifacts.",
},
{
"id": "single_episode_task_model_radar_json",
"title": "Single-episode 20-task model radar JSON",
"path": "docs/data/single_episode_task_model_radar.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Machine-readable split radar for the one-episode Minimal and Neural MLP baselines, both scored on all 20 task contracts.",
},
{
"id": "episode128_task_model_radar_json",
"title": "128-episode 20-task model radar JSON",
"path": "docs/data/episode128_task_model_radar.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Machine-readable split radar for selected 128-episode metadata/raw baselines, Qwen3-Omni v6, Cosmos3-Super, and Cosmos3-Nano, now complete at 140/140 scored rows with proxy notes retained.",
},
{
"id": "task_method_20_result_matrix_json",
"title": "Task-method 20-result matrix JSON",
"path": "docs/data/task_method_20_result_matrix.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Machine-readable 9-method by 20-task matrix where every method has 20 records and the current release is complete at 180/180 scored rows.",
},
{
"id": "task_method_20_result_matrix",
"title": "Task-method 20-result matrix",
"path": "TASK_METHOD_20_RESULT_MATRIX.md",
"kind": "evaluation_protocol",
"surface": "repo_hf",
"shows": "Reader-facing table that separates 20 records per method, direct numeric scores, documented compact-proxy scores, and source artifacts.",
},
{
"id": "task_method_20_gap_audit_json",
"title": "Task-method 20-result gap audit JSON",
"path": "docs/data/task_method_20_gap_audit.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Machine-readable 180-record completion ledger with numeric scores, proxy flags, explicit status reasons, and source artifacts.",
},
{
"id": "task_method_20_gap_audit",
"title": "Task-method 20-result gap audit",
"path": "TASK_METHOD_20_GAP_AUDIT.md",
"kind": "evaluation_protocol",
"surface": "repo_hf",
"shows": "Reader-facing ledger confirming 180/180 scored method-task cells and listing the six compact-proxy records separately.",
},
{
"id": "task_method_20_source_audit_json",
"title": "Task-method 20-result source audit JSON",
"path": "docs/data/task_method_20_source_audit.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Machine-readable check that scored JSON-backed matrix cells match their declared metric source values.",
},
{
"id": "task_method_20_source_audit",
"title": "Task-method 20-result source audit",
"path": "TASK_METHOD_20_SOURCE_AUDIT.md",
"kind": "evaluation_protocol",
"surface": "repo_hf",
"shows": "Reader-facing source-value audit for the 180-result matrix.",
},
{
"id": "two_evidence_line_map_chart",
"title": "Two evidence-line map",
"path": "docs/assets/charts/two_evidence_line_map.svg",
"kind": "generated_figure",
"surface": "website_hf",
"shows": "Explains the public result organization: one sample-episode task-lab line, one selected-128 comparison line, and the combined 180/180 scored method-task ledger.",
},
{
"id": "unified_task_model_radar_chart",
"title": "Unified 20-task model radar",
"path": "docs/assets/charts/unified_task_model_radar.svg",
"kind": "generated_figure",
"surface": "website_hf",
"shows": "Groups all nine methods into small-multiple 20-task radar panels so single-episode, 128-episode metadata/text, 128-episode raw-feature, and foundation-model rows remain readable.",
},
{
"id": "single_episode_task_model_radar_chart",
"title": "Single-episode 20-task model radar",
"path": "docs/assets/charts/single_episode_task_model_radar.svg",
"kind": "generated_figure",
"surface": "website_hf",
"shows": "Shows the one-episode Minimal and Neural MLP 20/20 scored baselines in one enlarged radar panel with local legend and task key.",
},
{
"id": "episode128_task_model_radar_chart",
"title": "128-episode 20-task model radar",
"path": "docs/assets/charts/episode128_task_model_radar.svg",
"kind": "generated_figure",
"surface": "website_hf",
"shows": "Separates selected 128-episode methods into metadata/text, raw-feature, and foundation-model radar panels with all 140 result rows scored and proxy notes retained.",
},
{
"id": "unified_task_model_radar_builder",
"title": "Unified 20-task model radar builder",
"path": "scripts/build_unified_task_model_radar.py",
"kind": "visualization_builder",
"surface": "repo_hf",
"shows": "Regenerates grouped 20-task radar charts plus machine-readable metric, source, chart-design, and proxy metadata.",
},
{
"id": "task_method_20_gap_audit_builder",
"title": "Task-method gap-audit builder",
"path": "scripts/build_task_method_20_gap_audit.py",
"kind": "publication_workflow",
"surface": "repo_hf",
"shows": "Regenerates the public completion/proxy audit from the 9-method by 20-task matrix without inventing unsupported scores.",
},
{
"id": "task_method_20_source_audit_validator",
"title": "Task-method source-audit validator",
"path": "scripts/validate_task_method_matrix_sources.py",
"kind": "publication_workflow",
"surface": "repo_hf",
"shows": "Fails release checks if a scored matrix row disagrees with its JSON metric source.",
},
{
"id": "all_task_model_scoring_waiter",
"title": "All-task model scoring guarded waiter",
"path": "scripts/omni/launch_all_task_model_scoring_when_free.sh",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Launches a user-provided all-task model scoring command only after enough private GPU capacity is idle, writing status logs under results/omni_finetune/deferred_launchers.",
},
{
"id": "model_output_probe_readiness",
"title": "Model-output probe readiness",
"path": "results/omni_finetune/model_output_probe_readiness/model_output_probe_readiness.json",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Checks whether Qwen3-Omni and Cosmos3 runs have train, validation, and test prediction files before extending model overlays to all 20 task contracts.",
},
{
"id": "model_output_probe_script",
"title": "Model-output probe readiness script",
"path": "scripts/omni/score_model_output_probes.py",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Audits model-output split availability and writes a readiness report without assigning new numeric task scores.",
},
{
"id": "existing_model_output_task_probe",
"title": "Existing model-output task probe package",
"path": "results/omni_finetune/model_output_task_probes_20260616/summary.json",
"kind": "model_result",
"surface": "repo_hf",
"shows": "Scores task-specific Qwen3-Omni and Cosmos3 overlays only where verified held-out prediction JSON or compact target maps already contain the required targets.",
},
{
"id": "existing_model_output_task_probe_script",
"title": "Existing model-output task probe scorer",
"path": "scripts/omni/score_existing_model_output_task_probes.py",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Derives task-specific scores from committed verified model outputs without running new inference or backfilling absent targets.",
},
{
"id": "a100_128_metadata_task_baselines",
"title": "128-episode metadata task baselines",
"path": "results/omni_finetune/a100_128_metadata_task_baselines_20260616_v2/summary_report.json",
"kind": "model_result",
"surface": "repo_hf",
"shows": "Rerun of JSONL metadata/text simple and neural baselines over the selected 128-episode multiscale dataset; supports radar overlays on JSONL-supported task axes.",
},
{
"id": "a100_128_raw20_task_baselines",
"title": "128-episode raw-feature 20-task baselines",
"path": "results/omni_finetune/a100_128_raw20_task_baselines_complete20_proxy_20260616T091500Z/run_summary_all.json",
"kind": "model_result",
"surface": "repo_hf",
"shows": "Rerun of simple and neural baselines over 34,269 windows and staged 4430-dimensional sensor NPZ features; covers 20 of 20 task axes, with interaction text and camera-view sync marked as compact-proxy completions because the 128 export lacks raw interaction strings and paired video-view embeddings.",
},
{
"id": "research_takeaways",
"title": "Research takeaways",
"path": "RESEARCH_TAKEAWAYS.md",
"kind": "result_interpretation",
"surface": "repo_hf",
"shows": "Summarizes the main research lessons from committed metrics and identifies which experiments need held-out episodes.",
},
{
"id": "research_takeaways_json",
"title": "Research takeaways JSON",
"path": "docs/data/research_takeaways.json",
"kind": "result_interpretation",
"surface": "website_hf",
"shows": "Machine-readable result interpretation for the website, HF cards, and mirror checks.",
},
{
"id": "research_takeaways_builder",
"title": "Research takeaways builder",
"path": "scripts/build_research_takeaways.py",
"kind": "result_interpretation",
"surface": "repo_hf",
"shows": "Regenerates the research takeaways from committed summary metrics and task result artifacts.",
},
{
"id": "audio_ablation_script",
"title": "Audio contribution script",
"path": "scripts/audio_ablation_and_raw_upgrade.py",
"kind": "result_interpretation",
"surface": "repo_hf",
"shows": "Measures audio contribution variants across the walkthrough-backed task contracts.",
},
{
"id": "audio_ablation_summary",
"title": "Audio ablation summary",
"path": "results/audio_ablation/audio_ablation_summary.json",
"kind": "metrics_source",
"surface": "repo_hf",
"shows": "Stores per-task audio deltas for all current features, no-audio, audio-only, alternate-audio-only, replacement, and all-plus-alternate variants.",
},
{
"id": "audio_ablation_summary_md",
"title": "Audio ablation summary report",
"path": "results/audio_ablation/AUDIO_ABLATION_SUMMARY.md",
"kind": "result_interpretation",
"surface": "repo_hf",
"shows": "Human-readable table showing the measured audio contribution and alternate-representation delta for every task.",
},
{
"id": "audio_ablation_website_json",
"title": "Audio ablation website JSON",
"path": "docs/data/audio_ablation_summary.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Machine-readable audio ablation summary mirrored into the static website and Hugging Face bundles.",
},
{
"id": "audio_ablation_delta_chart",
"title": "Audio ablation delta chart",
"path": "docs/assets/charts/audio_ablation_delta.svg",
"kind": "visual_evidence",
"surface": "website_hf",
"shows": "Bar chart of measured current-audio primary-metric deltas across the walkthrough-backed tasks.",
},
{
"id": "figure_index",
"title": "Figure index",
"path": "FIGURE_INDEX.md",
"kind": "visual_evidence",
"surface": "repo_hf",
"shows": "Catalogs public figures, charts, modality thumbnails, dimensions, hashes, roles, and source scripts.",
},
{
"id": "figure_index_json",
"title": "Figure index JSON",
"path": "docs/data/figure_index.json",
"kind": "visual_evidence",
"surface": "website_hf",
"shows": "Machine-readable visual asset index for website and Hugging Face mirrors.",
},
{
"id": "figure_index_builder",
"title": "Figure index builder",
"path": "scripts/build_figure_index.py",
"kind": "visual_evidence",
"surface": "repo_hf",
"shows": "Regenerates visual-asset hashes, dimensions, and source-script provenance.",
},
{
"id": "brand_assets_json",
"title": "Brand assets manifest",
"path": "docs/data/brand_assets.json",
"kind": "visual_evidence",
"surface": "website_hf",
"shows": "Machine-readable manifest for the generated logo system, favicon, social card, dimensions, hashes, and usage roles.",
},
{
"id": "brand_logo_social_card",
"title": "Brand logo social card",
"path": "docs/assets/brand/xperience10m-logo-social-card.png",
"kind": "visual_evidence",
"surface": "website_hf",
"shows": "Provides the project logo card used in README, Hugging Face cards, and social previews.",
},
{
"id": "brand_asset_builder",
"title": "Brand asset builder",
"path": "scripts/build_brand_assets.py",
"kind": "visual_evidence",
"surface": "repo_hf",
"shows": "Regenerates logo derivatives, favicon variants, app icons, and the Open Graph social card from the generated logo mark.",
},
{
"id": "raw_sample_files_manifest",
"title": "Raw public sample file manifest",
"path": "docs/data/raw_sample_files.json",
"kind": "dataset_context",
"surface": "website_hf",
"shows": "Lists the official public sample HDF5, MP4, and RRD files, derived browser-preview clips, playback/download URLs, file sizes, browser behavior, and HDF5 group organization.",
},
{
"id": "quality_gates",
"title": "Release checks",
"path": "QUALITY_GATES.md",
"kind": "quality_gate",
"surface": "repo_hf",
"shows": "Lists the automated and post-publish checks used to keep the release current.",
},
{
"id": "quality_gate_manifest",
"title": "Release-check manifest",
"path": "docs/data/quality_gates.json",
"kind": "quality_gate",
"surface": "website_hf",
"shows": "Machine-readable release-check summary for validators, mirrors, and public project surfaces.",
},
{
"id": "public_surface_qa",
"title": "Public project surface",
"path": "PUBLIC_SURFACE_QA.md",
"kind": "quality_gate",
"surface": "repo_hf",
"shows": "Keeps the repo, website, and Hugging Face cards aligned as one cohesive research project surface.",
},
{
"id": "public_reader_map",
"title": "Public reader map",
"path": "PUBLIC_READER_MAP.md",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Provides the first-pass navigation layer for GitHub, GitHub Pages, Hugging Face mirrors, Qwen3-Omni/Cosmos3 repos, evidence lines, and result-reading lanes.",
},
{
"id": "public_reader_map_json",
"title": "Public reader map JSON",
"path": "docs/data/public_reader_map.json",
"kind": "project_path",
"surface": "website_hf",
"shows": "Machine-readable public reader map used by the website and Hugging Face mirrors to keep entry points and surface responsibilities explicit.",
},
{
"id": "public_surface_qa_json",
"title": "Public project surface JSON",
"path": "docs/data/public_surface_qa.json",
"kind": "quality_gate",
"surface": "website_hf",
"volatile": True,
"shows": "Machine-readable report for SEO/social metadata, accessible tab semantics, public links, project links, and clear project presentation.",
},
{
"id": "public_surface_qa_builder",
"title": "Public project surface builder",
"path": "scripts/build_public_surface_qa.py",
"kind": "quality_gate",
"surface": "repo_hf",
"shows": "Regenerates the public presentation report before release.",
},
{
"id": "task_surface_integrity",
"title": "Task-surface integrity report",
"path": "docs/data/task_surface_integrity.json",
"kind": "quality_gate",
"surface": "website_hf",
"volatile": True,
"shows": "Confirms the public original-task cards use human-readable research names, representative modality thumbnails, and the interactive walkthrough/player JSON contract.",
},
{
"id": "rendered_site_check",
"title": "Rendered website check",
"path": "RENDERED_SITE_CHECK.md",
"kind": "quality_gate",
"surface": "repo_hf",
"volatile": True,
"shows": "Records the latest browser-level load, tab, walkthrough deep-link, control-click, and console-health check.",
},
{
"id": "rendered_site_check_json",
"title": "Rendered website check JSON",
"path": "docs/data/rendered_site_check.json",
"kind": "quality_gate",
"surface": "website_hf",
"volatile": True,
"shows": "Machine-readable browser-level website check for the public static site.",
},
{
"id": "rendered_site_check_builder",
"title": "Rendered website check builder",
"path": "scripts/build_rendered_site_check.py",
"kind": "quality_gate",
"surface": "repo_hf",
"shows": "Builds the rendered website check from browser observations.",
},
{
"id": "task_surface_validator",
"title": "Task-surface integrity validator",
"path": "scripts/validate_task_surface.py",
"kind": "quality_gate",
"surface": "repo_hf",
"shows": "Regenerates the task-surface integrity report and fails if task cards expose raw artifact ids or lose the interactive player wiring.",
},
{
"id": "live_publication_status",
"title": "Live publication status",
"path": "docs/data/live_publication_status.json",
"kind": "quality_gate",
"surface": "website_hf",
"volatile": True,
"shows": "Records the last live GitHub/HF URL verification after upload.",
},
{
"id": "live_publication_verifier",
"title": "Live publication verifier",
"path": "scripts/verify_live_publication.py",
"kind": "quality_gate",
"surface": "repo",
"shows": "Fetches the published GitHub/HF URLs and compares live hashes and public-card markers against the release assets.",
},
{
"id": "reproducibility_contract",
"title": "Reproducibility contract",
"path": "REPRODUCIBILITY.md",
"kind": "reproducibility",
"surface": "repo_hf",
"shows": "Defines public reproduction commands, expected outputs, and non-reproducible scale-up boundaries.",
},
{
"id": "reproducibility_matrix",
"title": "Reproducibility matrix",
"path": "docs/data/reproducibility_matrix.json",
"kind": "reproducibility",
"surface": "website_hf",
"shows": "Machine-readable reproduction steps with expected artifacts and public boundaries.",
},
{
"id": "artifact_index_builder",
"title": "Artifact index builder",
"path": "scripts/build_artifact_index.py",
"kind": "project_path",
"surface": "repo_hf",
"shows": "Generates the selective artifact catalog from local files.",
},
{
"id": "publication_audit",
"title": "Public bundle contents",
"path": "docs/data/publication_audit.json",
"kind": "publication_package_check",
"surface": "website_hf",
"volatile": True,
"shows": "Confirms public bundles exclude raw data, caches, heavy archives, and credential text.",
},
{
"id": "scale_up_status_check",
"title": "Multi-episode pilot status",
"path": "docs/data/scope_claims_audit.json",
"kind": "scale_up_status",
"surface": "website_hf",
"volatile": True,
"shows": "Separates setup paths from completed held-out-episode results.",
},
{
"id": "mirror_parity",
"title": "Prepared mirror parity report",
"path": "docs/data/mirror_parity.json",
"kind": "mirror_parity",
"surface": "website_hf",
"volatile": True,
"shows": "Confirms prepared GitHub/HF Space/artifact/model mirrors share the same critical data, figure, website HTML, and validator files.",
},
{
"id": "website_integrity",
"title": "Website integrity report",
"path": "docs/data/website_integrity.json",
"kind": "integrity_report",
"surface": "website_hf",
"volatile": True,
"shows": "Confirms local website links, anchors, JSON data files, and referenced images resolve.",
},
{
"id": "project_manifest",
"title": "Project manifest",
"path": "docs/data/project_manifest.json",
"kind": "metadata",
"surface": "website_hf",
"shows": "Lists public URLs, upstream sources, and machine-readable project metadata.",
},
{
"id": "task_summary",
"title": "Original task summary report",
"path": "results/episode_task_suite/summary_report.json",
"kind": "metrics_source",
"surface": "repo_hf",
"shows": "Stores the task definitions, splits, feature dimension, and minimal/neural metrics.",
},
{
"id": "website_metrics_bundle",
"title": "Website metrics bundle",
"path": "docs/data/summary_metrics.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Mirrors task metrics for the static dashboard.",
},
{
"id": "feature_manifest",
"title": "Feature manifest",
"path": "results/episode_task_suite/feature_manifest.json",
"kind": "data_contract",
"surface": "repo_hf",
"shows": "Maps the current window vector back to source feature blocks.",
},
{
"id": "available_modalities",
"title": "Available modalities",
"path": "results/episode_task_suite/available_modalities.json",
"kind": "data_contract",
"surface": "repo_hf",
"shows": "Documents which sample modalities entered the current extracted feature contract.",
},
{
"id": "windows_table",
"title": "Aligned windows table",
"path": "results/episode_task_suite/windows.csv",
"kind": "data_contract",
"surface": "repo_hf",
"shows": "Lists the 1,161 aligned windows and their frame/action/subtask labels.",
},
{
"id": "neural_mlp_directory",
"title": "Neural MLP task-head results",
"path": "results/episode_task_suite/neural_mlp",
"kind": "result_directory",
"surface": "repo_hf_model",
"shows": "Stores matching PyTorch MLP results for the walkthrough-backed task contracts.",
},
{
"id": "research_direction_taxonomy",
"title": "Research direction taxonomy",
"path": "results/episode_task_suite/research_directions/research_direction_taxonomy.json",
"kind": "taxonomy",
"surface": "repo_hf",
"shows": "Maps the walkthrough-backed tasks to the four Ropedia research directions as direct/proxy/diagnostic.",
},
{
"id": "research_direction_extensions",
"title": "Research direction extension probes",
"path": "results/episode_task_suite/research_direction_extensions/research_direction_extension_results.json",
"kind": "metrics_source",
"surface": "repo_hf",
"shows": "Stores one coded extension probe per research direction with minimal and neural metrics.",
},
{
"id": "tier2_task_suite",
"title": "Unified 20-task provenance bundle",
"path": "results/episode_task_suite/tier2_task_suite/tier2_task_suite_results.json",
"kind": "metrics_source",
"surface": "repo_hf",
"shows": "Stores the historical result bundle for provenance rows with minimal and neural baselines aligned to the same 20-task window/split setup.",
},
{
"id": "tier2_task_suite_json",
"title": "Unified 20-task provenance JSON",
"path": "docs/data/tier2_task_suite.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Machine-readable provenance definitions, setup alignment, metrics, and public source paths; the file name is historical.",
},
{
"id": "tier2_task_suite_chart",
"title": "Unified 20-task provenance chart",
"path": "docs/assets/charts/tier2_task_suite.svg",
"kind": "generated_figure",
"surface": "website_hf",
"shows": "Visual summary of the historical provenance baseline metrics inside the unified 20-task suite.",
},
{
"id": "tier2_task_suite_builder",
"title": "Unified 20-task provenance builder",
"path": "scripts/tier2_task_suite.py",
"kind": "evaluation_protocol",
"surface": "repo_hf",
"shows": "Regenerates the historical provenance rows from shared windows plus the local public-sample annotation HDF5; the script name is historical.",
},
{
"id": "task_walkthroughs",
"title": "Task walkthroughs",
"path": "results/episode_task_suite/task_walkthroughs/TASK_WALKTHROUGHS.md",
"kind": "onboarding_doc",
"surface": "repo_hf",
"shows": "Explains every task with case study, input, process modules, output, and limitation.",
},
{
"id": "task_suite_infographic",
"title": "Original task-suite infographic",
"path": "docs/assets/task_suite_infographic.png",
"kind": "generated_figure",
"surface": "website_hf",
"shows": "Presents the task suite and sample modality thumbnails with metrics generated from committed files.",
},
{
"id": "modality_atlas",
"title": "Responsive modality atlas",
"path": "docs/data/modality_atlas.json",
"kind": "website_data",
"surface": "website_hf",
"shows": "Documents the seven public-sample modality cards and their derived thumbnail assets.",
},
{
"id": "modality_thumbnails",
"title": "Standalone modality thumbnails",
"path": "docs/assets/modalities",
"kind": "generated_figure_assets",
"surface": "website_hf",
"shows": "Stores small derived thumbnails for readable website modality cards without raw data redistribution.",
},
{
"id": "pipeline_figure",
"title": "Pipeline figure",
"path": "docs/assets/pipeline_diagram.png",
"kind": "generated_figure",
"surface": "website_hf",
"shows": "Shows the raw-episode to artifact pipeline with verified labels.",
},
{
"id": "architecture_figure",
"title": "Architecture figure",
"path": "docs/assets/task_architectures.png",
"kind": "generated_figure",
"surface": "website_hf",
"shows": "Shows the shared feature pipeline and minimal/neural head families.",
},
{
"id": "qwen_data_access_status",
"title": "Qwen3-Omni data access status",
"path": "results/omni_finetune/DATA_ACCESS_STATUS.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Summarizes the data-readiness checks required before a held-out Qwen3-Omni pilot can report metrics.",
},
{
"id": "qwen3_lora_hf_upload_note",
"title": "Qwen3 LoRA HF upload note",
"path": "results/omni_finetune/HF_UPLOAD.md",
"kind": "publication_workflow",
"surface": "repo_hf",
"shows": "Documents the final 128-episode LoRA adapter upload path, target model repo, package builder, and forbidden files.",
},
{
"id": "multi_episode_access_status",
"title": "Multi-episode access status",
"path": "results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Documents the public multi-episode access status and 32-episode pilot selection.",
},
{
"id": "qwen3_omni_error_analysis_report",
"title": "Qwen3-Omni held-out error-analysis report",
"path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605_eval/analysis/ERROR_ANALYSIS.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Summarizes the earlier validation-aware Qwen3-Omni held-out failures by episode, action family, train-seen status, required-modality state, and object category.",
},
{
"id": "qwen3_omni_error_analysis_json",
"title": "Qwen3-Omni held-out error-analysis JSON",
"path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605_eval/analysis/error_analysis_summary.json",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Machine-readable Qwen3-Omni held-out error analysis with grouped metrics and sanitized failure examples.",
},
{
"id": "multi_episode_128_baseline_report",
"title": "128-episode aligned baseline report",
"path": "results/omni_finetune/multi_episode_128_task_baselines/BASELINE_ALIGNMENT_REPORT.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Summarizes same-split simple and neural metadata baselines for the walkthrough-backed task ids, with unsupported markers for tasks that need missing raw 128 feature blocks.",
},
{
"id": "multi_episode_128_baseline_summary",
"title": "128-episode aligned baseline summary",
"path": "results/omni_finetune/multi_episode_128_task_baselines/summary_report.json",
"kind": "metrics_source",
"surface": "repo_hf",
"shows": "Machine-readable 96/16/16 split counts, run configuration, per-task simple metrics, neural metrics, and raw-feature unsupported statuses.",
},
{
"id": "omni_model_comparison_report",
"title": "Omni model comparison report",
"path": "results/omni_finetune/OMNI_MODEL_COMPARISON.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Reader-facing comparison of the single-episode task suite, 128-episode aligned baselines, Qwen3-Omni packages, and Cosmos3 future-window branch.",
},
{
"id": "omni_model_comparison_json",
"title": "Omni model comparison JSON",
"path": "docs/data/omni_model_comparison.json",
"kind": "metrics_source",
"surface": "repo_hf",
"shows": "Machine-readable comparison of the current result versions, per-task aligned baselines, verified Qwen3 packages, and Cosmos3 package.",
},
{
"id": "cosmos3_nano_verified_summary",
"title": "Cosmos3-Nano verified package summary",
"path": "results/omni_finetune/verified_public/xperience10m_cosmos3_nano_128ep_future_window_h5_compat_adapter_eval_test_full/verified_result_summary.json",
"kind": "metrics_source",
"surface": "repo_hf",
"shows": "Machine-readable verified public summary for the Cosmos3-Nano future-window compatibility package.",
},
{
"id": "cosmos3_nano_run_report",
"title": "Cosmos3-Nano future-window run report",
"path": "results/omni_finetune/verified_public/xperience10m_cosmos3_nano_128ep_future_window_h5_compat_adapter_eval_test_full/eval/RUN_REPORT.md",
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Reader-facing held-out metrics and interpretation for the Cosmos3-Nano future-window compatibility branch.",
},
{
"id": "citation",
"title": "Citation metadata",
"path": "CITATION.cff",
"kind": "citation",
"surface": "repo_hf",
"shows": "Makes the project externally citable.",
},
{
"id": "license",
"title": "License and data terms",
"path": "LICENSE",
"kind": "license",
"surface": "repo_hf",
"shows": "Separates MIT-scoped code from original Xperience-10M data terms.",
},
]
def sha256(path: Path) -> str:
digest = hashlib.sha256()
with path.open("rb") as handle:
for chunk in iter(lambda: handle.read(1024 * 1024), b""):
digest.update(chunk)
return digest.hexdigest()
def directory_stats(path: Path) -> dict:
files = [item for item in path.rglob("*") if item.is_file()]
return {
"file_count": len(files),
"bytes": sum(item.stat().st_size for item in files),
}
def verified_public_package_artifacts() -> list[dict]:
verified_root = ROOT / "results/omni_finetune/verified_public"
if not verified_root.exists():
return []
artifacts: list[dict] = []
for summary_path in sorted(verified_root.glob("*/verified_result_summary.json")):
package_dir = summary_path.parent
slug = package_dir.name
payload = json.loads(summary_path.read_text(encoding="utf-8"))
title = payload.get("backbone_display_name") or payload.get("eval_run_id") or slug
backbone = payload.get("backbone", "unknown_backbone")
status = payload.get("status", "unknown")
eval_run_id = payload.get("eval_run_id", slug)
artifacts.append(
{
"id": f"verified_public_package_{slug}",
"title": f"Verified public package: {title}",
"path": package_dir.relative_to(ROOT).as_posix(),
"kind": "verified_public_package",
"surface": "repo_hf",
"shows": (
f"Public-safe verified package for {eval_run_id} "
f"({backbone}, status={status})."
),
}
)
artifacts.append(
{
"id": f"verified_public_summary_{slug}",
"title": f"Verified summary: {title}",
"path": summary_path.relative_to(ROOT).as_posix(),
"kind": "metrics_source",
"surface": "repo_hf",
"shows": f"Machine-readable verified summary for {eval_run_id}.",
}
)
for relative_path, kind, label in [
("PUBLIC_RESULT_SUMMARY.md", "scaleup_status", "public result summary"),
("eval/RUN_REPORT.md", "scaleup_status", "run report"),
("eval/metrics.json", "metrics_source", "metrics JSON"),
("package_audit.json", "publication_audit", "package audit"),
]:
path = package_dir / relative_path
if not path.exists():
continue
safe_label = label.replace(" ", "_")
artifacts.append(
{
"id": f"verified_public_{safe_label}_{slug}",
"title": f"Verified {label}: {title}",
"path": path.relative_to(ROOT).as_posix(),
"kind": kind,
"surface": "repo_hf",
"shows": f"{label.capitalize()} for {eval_run_id}.",
}
)
return artifacts
def qwen3_future_task_probe_artifacts() -> list[dict]:
run_dir = ROOT / "results/omni_finetune" / QWEN3_FUTURE_TASK_PROBE_RUN_ID
if not run_dir.exists():
return []
artifacts: list[dict] = [
{
"id": "qwen3_future_task_probe_package",
"title": "Qwen3 v6 future-task probe package",
"path": run_dir.relative_to(ROOT).as_posix(),
"kind": "model_result",
"surface": "repo_hf",
"shows": (
"Two-shard Qwen3-Omni v6 inference probe for tasks 13, 14, "
"and 17, with public-safe metrics, predictions, progress logs, "
"and merge report."
),
}
]
for relative_path, kind, label in [
("summary.json", "metrics_source", "merged summary"),
("collection_validation.json", "publication_audit", "collection validation"),
("RUN_REPORT.md", "scaleup_status", "run report"),
("long_horizon_next_action/metrics.json", "metrics_source", "task 13 metrics"),
("next_subtask_forecast/metrics.json", "metrics_source", "task 14 metrics"),
("object_set_forecast/metrics.json", "metrics_source", "task 17 metrics"),
]:
path = run_dir / relative_path
if path.exists():
artifacts.append(
{
"id": f"qwen3_future_task_probe_{relative_path.replace('/', '_').replace('.', '_')}",
"title": f"Qwen3 future-task probe {label}",
"path": path.relative_to(ROOT).as_posix(),
"kind": kind,
"surface": "repo_hf",
"shows": f"Public-safe {label} for {QWEN3_FUTURE_TASK_PROBE_RUN_ID}.",
}
)
launcher_log = (
ROOT
/ "results/omni_finetune/deferred_launchers"
/ f"{QWEN3_FUTURE_TASK_PROBE_RUN_ID}.launcher.log"
)
if launcher_log.exists():
artifacts.append(
{
"id": "qwen3_future_task_probe_launcher_log",
"title": "Qwen3 future-task probe launcher log",
"path": launcher_log.relative_to(ROOT).as_posix(),
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Launch and merge log for the two-shard future-task probe.",
}
)
return artifacts
def cosmos3_super_interaction_text_probe_artifacts() -> list[dict]:
run_dir = ROOT / "results/omni_finetune" / COSMOS3_SUPER_INTERACTION_TEXT_TASK_PROBE_RUN_ID
if not run_dir.exists():
return []
artifacts: list[dict] = [
{
"id": "cosmos3_super_interaction_text_probe_package",
"title": "Cosmos3-Super interaction-text task-15 probe package",
"path": run_dir.relative_to(ROOT).as_posix(),
"kind": "model_result",
"surface": "repo_hf",
"shows": (
"Four-shard Cosmos3-Super text-only inference probe for task 15 "
"over raw annotation.hdf5 interaction text labels."
),
}
]
for relative_path, kind, label in [
("summary.json", "metrics_source", "merged summary"),
("launch_env.txt", "scaleup_status", "launch environment"),
("interaction_text_prediction/RUN_REPORT.md", "scaleup_status", "task 15 run report"),
("interaction_text_prediction/metrics.json", "metrics_source", "task 15 metrics"),
("interaction_text_prediction/per_class_metrics.csv", "metrics_source", "task 15 per-class metrics"),
("interaction_text_prediction/confusion_matrix.csv", "metrics_source", "task 15 confusion matrix"),
]:
path = run_dir / relative_path
if path.exists():
artifacts.append(
{
"id": f"cosmos3_super_interaction_text_{relative_path.replace('/', '_').replace('.', '_')}",
"title": f"Cosmos3-Super interaction-text probe {label}",
"path": path.relative_to(ROOT).as_posix(),
"kind": kind,
"surface": "repo_hf",
"shows": f"Public-safe {label} for {COSMOS3_SUPER_INTERACTION_TEXT_TASK_PROBE_RUN_ID}.",
}
)
launcher_log = (
ROOT
/ "results/omni_finetune/deferred_launchers"
/ f"{COSMOS3_SUPER_INTERACTION_TEXT_TASK_PROBE_RUN_ID}.launcher.log"
)
if launcher_log.exists():
artifacts.append(
{
"id": "cosmos3_super_interaction_text_probe_launcher_log",
"title": "Cosmos3-Super interaction-text probe launcher log",
"path": launcher_log.relative_to(ROOT).as_posix(),
"kind": "scaleup_status",
"surface": "repo_hf",
"shows": "Launch and merge log for the four-shard task-15 probe.",
}
)
return artifacts
def artifact_entry(item: dict) -> dict:
path = ROOT / item["path"]
entry = {
**item,
"exists": path.exists(),
}
if path.is_file():
entry["bytes"] = path.stat().st_size
if item.get("volatile"):
entry["hash_policy"] = "existence_and_size_only"
else:
entry["sha256"] = sha256(path)
elif path.is_dir():
entry.update(directory_stats(path))
else:
entry.update({"bytes": 0})
return entry
def main() -> int:
artifacts = [dict(item) for item in ARTIFACTS]
artifacts.extend(verified_public_package_artifacts())
artifacts.extend(qwen3_future_task_probe_artifacts())
artifacts.extend(cosmos3_super_interaction_text_probe_artifacts())
summary_path = ROOT / "results/episode_task_suite/summary_report.json"
if summary_path.exists():
summary = json.loads(summary_path.read_text(encoding="utf-8"))
feature_dim = int(summary.get("feature_dim", 0))
for item in artifacts:
if item["id"] == "feature_manifest" and feature_dim:
item["shows"] = f"Maps the {feature_dim:,}-dimensional window vector back to source feature blocks."
entries = [artifact_entry(item) for item in artifacts]
missing = [entry["path"] for entry in entries if not entry["exists"]]
by_kind: dict[str, int] = {}
for entry in entries:
by_kind[entry["kind"]] = by_kind.get(entry["kind"], 0) + 1
report = {
"title": "Ropedia Xperience-10M Task Suite Artifact Index",
"generated_at_utc": datetime.now(timezone.utc).isoformat(timespec="seconds"),
"status": "pass" if not missing else "fail",
"artifact_count": len(entries),
"missing": missing,
"by_kind": by_kind,
"artifacts": entries,
}
OUTPUT.parent.mkdir(parents=True, exist_ok=True)
OUTPUT.write_text(json.dumps(report, indent=2) + "\n", encoding="utf-8")
print(f"{report['status'].upper()}: wrote {OUTPUT}")
if missing:
for path in missing:
print(f"- missing: {path}")
return 1
return 0
if __name__ == "__main__":
raise SystemExit(main())
|