Robotics
PyTorch
Cosmos
xperience10m_task_baseline_suite
embodied-ai
multimodal
xperience-10m
baseline
evaluation
qwen3-omni
Instructions to use cy0307/ropedia-xperience-10m-task-baselines with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Cosmos
How to use cy0307/ropedia-xperience-10m-task-baselines with Cosmos:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
File size: 20,037 Bytes
13d3eec b158b4d 13d3eec b158b4d 13d3eec 557a248 13d3eec 557a248 13d3eec b158b4d 13d3eec b158b4d 13d3eec 67b10a0 13d3eec 67b10a0 d73afa7 13d3eec d73afa7 13d3eec 67b10a0 13d3eec 67b10a0 d73afa7 13d3eec d73afa7 13d3eec 17c38d5 13d3eec 17c38d5 13d3eec 557a248 b158b4d d73afa7 17c38d5 13d3eec b158b4d d73afa7 13d3eec 67b10a0 13d3eec b158b4d 13d3eec 557a248 13d3eec 7f5180a 13d3eec 67b10a0 13d3eec 67b10a0 13d3eec 17c38d5 13d3eec eeaf70e 13d3eec b158b4d 13d3eec a3e1c37 13d3eec 67b10a0 13d3eec eeaf70e 67b10a0 13d3eec eeaf70e 13d3eec 67b10a0 13d3eec 67b10a0 13d3eec 67b10a0 13d3eec b158b4d 13d3eec | 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 | {
"generated_at_utc": "2026-06-20T13:58:05+00:00",
"immediate_actions": [
{
"artifact": "docs/data/task_method_20_gap_audit.json",
"id": "gap_audit",
"purpose": "Keep the 16 scoreless cells visible and reproducible."
},
{
"artifact": "scripts/omni/score_model_output_probes.py",
"id": "model_output_probe",
"purpose": "Check whether train/validation/test model outputs exist before attempting all-task Qwen3/Cosmos scoring."
},
{
"artifact": "scripts/omni/launch_all_task_model_scoring_when_free.sh",
"id": "guarded_gpu_launcher",
"purpose": "Start a user-provided all-task scoring command only after enough private GPU capacity is idle."
}
],
"methods": {
"cosmos3_nano_future_window": {
"kind": "partial_128_episode_world_model_overlay",
"label": "Cosmos3-Nano Future Window",
"proxy_scored_task_count": 0,
"result_record_count": 20,
"scope": "128 selected episodes, held-out test",
"scored_task_count": 11,
"scoreless_task_count": 9,
"status_counts": {
"not_evaluated_in_verified_package": 9,
"scored": 11
}
},
"cosmos3_super_reasoner": {
"kind": "partial_128_episode_foundation_model_overlay",
"label": "Cosmos3-Super Reasoner",
"proxy_scored_task_count": 0,
"result_record_count": 20,
"scope": "128 selected episodes, held-out test",
"scored_task_count": 15,
"scoreless_task_count": 5,
"status_counts": {
"not_evaluated_in_verified_package": 5,
"scored": 15
}
},
"metadata128_neural_mlp": {
"kind": "partial_128_episode_aligned_baseline",
"label": "128ep Aligned NN",
"proxy_scored_task_count": 0,
"result_record_count": 20,
"scope": "128 selected episodes, JSONL metadata/text plus staged sensor-block targets where available",
"scored_task_count": 19,
"scoreless_task_count": 1,
"status_counts": {
"not_supported_by_metadata_only_package": 1,
"scored": 19
}
},
"metadata128_simple": {
"kind": "partial_128_episode_aligned_baseline",
"label": "128ep Aligned Simple",
"proxy_scored_task_count": 0,
"result_record_count": 20,
"scope": "128 selected episodes, JSONL metadata/text plus staged sensor-block targets where available",
"scored_task_count": 19,
"scoreless_task_count": 1,
"status_counts": {
"scored": 19,
"unsupported_without_required_target": 1
}
},
"minimal": {
"kind": "full_20_task_baseline",
"label": "Minimal",
"proxy_scored_task_count": 0,
"result_record_count": 20,
"scope": "1 public sample episode",
"scored_task_count": 20,
"scoreless_task_count": 0,
"status_counts": {
"scored": 20
}
},
"neural_mlp": {
"kind": "full_20_task_baseline",
"label": "Neural MLP",
"proxy_scored_task_count": 0,
"result_record_count": 20,
"scope": "1 public sample episode",
"scored_task_count": 20,
"scoreless_task_count": 0,
"status_counts": {
"scored": 20
}
},
"qwen3_omni_v6_lora": {
"kind": "partial_128_episode_foundation_model_overlay",
"label": "Qwen3-Omni v6 LoRA",
"proxy_scored_task_count": 0,
"result_record_count": 20,
"scope": "128 selected episodes, held-out test",
"scored_task_count": 20,
"scoreless_task_count": 0,
"status_counts": {
"scored": 20
}
},
"raw128_neural_mlp": {
"kind": "complete_128_episode_raw_feature_baseline",
"label": "128ep Raw NN",
"proxy_scored_task_count": 2,
"result_record_count": 20,
"scope": "128 selected episodes, staged 4430-dim sensor NPZ features; 2 compact proxy axes",
"scored_task_count": 20,
"scoreless_task_count": 0,
"status_counts": {
"proxy_scored": 2,
"scored": 18
}
},
"raw128_simple": {
"kind": "complete_128_episode_raw_feature_baseline",
"label": "128ep Raw Simple",
"proxy_scored_task_count": 2,
"result_record_count": 20,
"scope": "128 selected episodes, staged 4430-dim sensor NPZ features; 2 compact proxy axes",
"scored_task_count": 20,
"scoreless_task_count": 0,
"status_counts": {
"proxy_scored": 2,
"scored": 18
}
}
},
"missing_by_method": {
"cosmos3_nano_future_window": 9,
"cosmos3_super_reasoner": 5,
"metadata128_neural_mlp": 1,
"metadata128_simple": 1
},
"missing_by_status": {
"not_evaluated_in_verified_package": 14,
"not_supported_by_metadata_only_package": 1,
"unsupported_without_required_target": 1
},
"missing_by_task": {
"02 Procedure Step Recognition": [
"cosmos3_nano_future_window"
],
"05 Hand Trajectory Forecasting": [
"cosmos3_nano_future_window"
],
"07 Object Relevance Prediction": [
"cosmos3_nano_future_window"
],
"08 Language Grounding": [
"cosmos3_nano_future_window"
],
"11 Temporal Order Verification": [
"cosmos3_nano_future_window",
"cosmos3_super_reasoner"
],
"12 Multimodal Synchronization Detection": [
"cosmos3_nano_future_window",
"cosmos3_super_reasoner"
],
"14 Long-Horizon Next-Subtask Forecasting": [
"cosmos3_super_reasoner"
],
"15 Interaction Text Prediction": [
"cosmos3_nano_future_window",
"cosmos3_super_reasoner"
],
"17 Future Object-Set Forecasting": [
"cosmos3_super_reasoner"
],
"18 IMU-to-Hand Pose Reconstruction": [
"cosmos3_nano_future_window"
],
"19 Camera-View Synchronization Retrieval": [
"cosmos3_nano_future_window",
"metadata128_neural_mlp",
"metadata128_simple"
]
},
"missing_records": [
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "macro_f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "timeline_subtask",
"task_label": "Procedure Step Recognition",
"task_number": 2
},
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "mpjpe",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "hand_trajectory_forecast",
"task_label": "Hand Trajectory Forecasting",
"task_number": 5
},
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "micro_f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "object_relevance",
"task_label": "Object Relevance Prediction",
"task_number": 7
},
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "mrr",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "caption_grounding",
"task_label": "Language Grounding",
"task_number": 8
},
{
"method": "Cosmos3-Super Reasoner",
"metric_key": "f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_super_reasoner",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "temporal_order",
"task_label": "Temporal Order Verification",
"task_number": 11
},
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "temporal_order",
"task_label": "Temporal Order Verification",
"task_number": 11
},
{
"method": "Cosmos3-Super Reasoner",
"metric_key": "f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_super_reasoner",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "misalignment_detection",
"task_label": "Multimodal Synchronization Detection",
"task_number": 12
},
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "misalignment_detection",
"task_label": "Multimodal Synchronization Detection",
"task_number": 12
},
{
"method": "Cosmos3-Super Reasoner",
"metric_key": "macro_f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_super_reasoner",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "next_subtask_forecast",
"task_label": "Long-Horizon Next-Subtask Forecasting",
"task_number": 14
},
{
"method": "Cosmos3-Super Reasoner",
"metric_key": "macro_f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_super_reasoner",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "interaction_text_prediction",
"task_label": "Interaction Text Prediction",
"task_number": 15
},
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "macro_f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "interaction_text_prediction",
"task_label": "Interaction Text Prediction",
"task_number": 15
},
{
"method": "Cosmos3-Super Reasoner",
"metric_key": "micro_f1",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_super_reasoner",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "object_set_forecast",
"task_label": "Future Object-Set Forecasting",
"task_number": 17
},
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "mae",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "imu_to_hand_pose",
"task_label": "IMU-to-Hand Pose Reconstruction",
"task_number": 18
},
{
"method": "128ep Aligned Simple",
"metric_key": "mrr",
"reason": "requires paired camera-view feature blocks, which are not in the public 128 JSONL metadata package",
"recommended_next_step": "Export the missing target field for this 128-episode method, then rerun the same train/validation/test split.",
"scope": "multi_episode_128_aligned_baseline",
"series_id": "metadata128_simple",
"status": "unsupported_without_required_target",
"status_label": "unsupported",
"task_id": "camera_view_sync_retrieval",
"task_label": "Camera-View Synchronization Retrieval",
"task_number": 19
},
{
"method": "128ep Aligned NN",
"metric_key": "mrr",
"reason": "the 128-episode aligned rerun did not produce this task target; raw interaction text, paired camera-view embeddings, or a task-specific target builder is required",
"recommended_next_step": "Run the task with raw sensor-feature blocks or add a task-specific metadata target builder before assigning a numeric score.",
"scope": "multi_episode_128_aligned_baseline",
"series_id": "metadata128_neural_mlp",
"status": "not_supported_by_metadata_only_package",
"status_label": "not supported",
"task_id": "camera_view_sync_retrieval",
"task_label": "Camera-View Synchronization Retrieval",
"task_number": 19
},
{
"method": "Cosmos3-Nano Future Window",
"metric_key": "mrr",
"reason": "the verified public model package did not ask this branch to emit that task target; a new task-specific evaluation package is required for a numeric score",
"recommended_next_step": "Generate verified model outputs for this task contract and score them against the held-out labels.",
"scope": "multi_episode_128_partial_model_overlay",
"series_id": "cosmos3_nano_future_window",
"status": "not_evaluated_in_verified_package",
"status_label": "not evaluated",
"task_id": "camera_view_sync_retrieval",
"task_label": "Camera-View Synchronization Retrieval",
"task_number": 19
}
],
"proxy_records": [
{
"method": "128ep Raw Simple",
"metric_key": "macro_f1",
"reason": "documented compact proxy completion for this raw128 task axis",
"series_id": "raw128_simple",
"source": "results/omni_finetune/a100_128_raw20_task_baselines_complete20_proxy_20260616T091500Z/simple_raw128/interaction_text_prediction/metrics.json",
"task_id": "interaction_text_prediction",
"task_label": "Interaction Text Prediction",
"task_number": 15
},
{
"method": "128ep Raw NN",
"metric_key": "macro_f1",
"reason": "documented compact proxy completion for this raw128 task axis",
"series_id": "raw128_neural_mlp",
"source": "results/omni_finetune/a100_128_raw20_task_baselines_complete20_proxy_20260616T091500Z/neural_mlp_raw128/interaction_text_prediction/metrics.json",
"task_id": "interaction_text_prediction",
"task_label": "Interaction Text Prediction",
"task_number": 15
},
{
"method": "128ep Raw Simple",
"metric_key": "mrr",
"reason": "documented compact proxy completion for this raw128 task axis",
"series_id": "raw128_simple",
"source": "results/omni_finetune/a100_128_raw20_task_baselines_complete20_proxy_20260616T091500Z/simple_raw128/camera_view_sync_retrieval/metrics.json",
"task_id": "camera_view_sync_retrieval",
"task_label": "Camera-View Synchronization Retrieval",
"task_number": 19
},
{
"method": "128ep Raw NN",
"metric_key": "mrr",
"reason": "documented compact proxy completion for this raw128 task axis",
"series_id": "raw128_neural_mlp",
"source": "results/omni_finetune/a100_128_raw20_task_baselines_complete20_proxy_20260616T091500Z/neural_mlp_raw128/camera_view_sync_retrieval/metrics.json",
"task_id": "camera_view_sync_retrieval",
"task_label": "Camera-View Synchronization Retrieval",
"task_number": 19
}
],
"score_summary": {
"method_count": 9,
"method_task_record_count": 180,
"proxy_scored_method_task_count": 4,
"scored_method_task_count": 164,
"scoreless_method_task_count": 16,
"task_count": 20
},
"source_matrix": "docs/data/task_method_20_result_matrix.json",
"status": "pass",
"target_policy": {
"numeric_score_gate": "A method-task cell is numeric only when a runner or verified package emits that exact task target and metric.",
"proxy_policy": "Proxy scores are allowed only when the matrix marks them as proxy_scored and keeps the reason/source attached.",
"scoreless_cell_policy": "Unsupported and not-evaluated cells stay explicit in the public matrix instead of being hidden or backfilled with proxy model claims."
},
"title": "Task Method 20-Result Gap Audit"
}
|