Cross-Embodiment Causal Transfer from Human Videos to Robot Manipulation
Scaling robot manipulation models requires large amounts of interaction data, yet collecting diverse real-robot experience remains costly.
ProofPaper ↗
Key points
- Human manipulation videos, in contrast, provide abundant and diverse task experience that could potentially benefit robot learning.
- However, differences in morphology, action spaces, and control mechanisms make human motions unsuitable as direct supervision for robot actions, raising a scientific question: what knowledge in human videos should be transferred across embodiments to robot manipulation?
- Based on this perspective, we propose Causal-Beam-Guided Subtask Transfer (CBST).
- CBST learns manipulation-related representations from human videos and grounds them in task-level semantic variables using robot demonstrations.
Sources (1)
- [1]Cross-Embodiment Causal Transfer from Human Videos to Robot ManipulationarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 15, 12:48 PM
Scaling robot manipulation models requires large amounts of interaction data, yet collecting diverse real-robot experience remains costly.
Human manipulation videos, in contrast, provide abundant and diverse task experience that could potentially benefit robot learning.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 15, 2026Decomposing Brain-LLM Alignment with Interpretable Semantic Representations
- Oct 15, 2026Hard Way, The Only Way: Provable Merits of Training on Hardest Examples
- Oct 15, 2026Measuring Mechanistic Uncertainty in Large Language Models via Causal Robustness
- Oct 15, 2026Lexical Reference Frames: Interpreting LM Residual Streams in Native Semantics
- Oct 14, 2026Empirical Boundaries of Reservoir Computing for Few-Shot Sequence Pattern Tasks