| { |
| "title": "Ropedia Xperience-10M Project Brief", |
| "summary": "A concise first-reader brief for the public-sample embodied-AI task lab and its multi-episode scale-up path.", |
| "research_intent": "Treat the public Xperience-10M sample as a small but real research system: inspect synchronized modalities, define embodied-AI task contracts, run bounded baselines, package evidence, and prepare held-out multi-episode scale-up without presenting one episode as final model quality.", |
| "capability_map": [ |
| { |
| "capability": "Data understanding", |
| "evidence": "feature_manifest.json, available_modalities.json, modality atlas, and the episode-window HF viewer" |
| }, |
| { |
| "capability": "Task design", |
| "evidence": "12 task contracts, task cards, case-study walkthroughs, and four research-direction extension probes" |
| }, |
| { |
| "capability": "Evaluation rigor", |
| "evidence": "chronological split, per-task metrics, predictions, confusion matrices, leakage notes, and generated takeaways" |
| }, |
| { |
| "capability": "Scale-up planning", |
| "evidence": "128-episode selection plan, Qwen3-Omni path, Cosmos 3 branch, and policy-model candidates after action-space conversion" |
| } |
| ], |
| "current_artifacts": [ |
| { |
| "layer": "Data unit", |
| "status": "1 public sample episode, 5,821 frames, 1,161 synchronized 20-frame windows" |
| }, |
| { |
| "layer": "Modalities", |
| "status": "Video-derived features, audio, depth, pose/SLAM, mocap, IMU, calibration, and language-derived features" |
| }, |
| { |
| "layer": "Task suite", |
| "status": "12 embodied-AI task contracts with inputs, targets, metrics, predictions, and case-study walkthroughs" |
| }, |
| { |
| "layer": "Models", |
| "status": "Minimal linear/ridge/logistic baselines plus compact PyTorch MLP heads for the same 12 tasks" |
| }, |
| { |
| "layer": "Research map", |
| "status": "Four Ropedia research directions with direct, proxy, diagnostic, and extension-task coverage" |
| }, |
| { |
| "layer": "Scale-up path", |
| "status": "Qwen3-Omni LoRA code path prepared; the gated full dataset is available for selected 128-episode data preparation" |
| } |
| ], |
| "reading_order": [ |
| "Start with the website or PROJECT_BRIEF.md to understand the project shape.", |
| "Open RESEARCH_ROADMAP.md to see how the work scales from the public sample to multi-episode modeling.", |
| "Open EVALUATION_PROTOCOL.md before comparing task scores.", |
| "Use RESEARCH_TAKEAWAYS.md for the current metric interpretation.", |
| "Inspect results/episode_task_suite/feature_manifest.json to understand one model input.", |
| "Use results/omni_finetune/DATA_ACCESS_STATUS.md for the multi-episode data status." |
| ], |
| "scope_boundary": "The public sample is enough to build and verify task definitions, feature contracts, metrics, visualization, and baseline code. It is not enough to measure final model quality for a general embodied-AI model.", |
| "next_stage": "Run the same contracts on held-out episodes, then fine-tune and evaluate an omni-model with train/test separation at the episode level.", |
| "entry_points": { |
| "visual_dashboard": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/", |
| "hf_space": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite", |
| "artifact_dataset": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts", |
| "baseline_model_bundle": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines", |
| "official_xperience10m_dataset": "https://huggingface.co/datasets/ropedia-ai/xperience-10m" |
| } |
| } |
|
|
