# Human2Robot: Learning Robot Actions from Paired Human-Robot Videos

arXiv: https://arxiv.org/abs/2502.16587 · HTML: https://mlsanity.com/paper/2502.16587/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Feb 2025.

## Built on by (most cited first, 25 of 25)

- [Scalable Vision-Language-Action Model Pretraining for Robotic Manipulation with Real-Life Human Activity Videos](https://mlsanity.com/paper/2510.21571/) (2025)
- [MagicMotion: Controllable Video Generation with Dense-to-Sparse Trajectory Guidance](https://mlsanity.com/paper/2503.16421/) (2025)
- [Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisons](https://mlsanity.com/paper/2603.02115/) (2026)
- [World Action Models: The Next Frontier in Embodied AI](https://mlsanity.com/paper/2605.12090/) (2026)
- [H-RDT: Human Manipulation Enhanced Bimanual Robotic Manipulation](https://mlsanity.com/paper/2507.23523/) (2025)
- [MimicDreamer: Aligning Human and Robot Demonstrations for Scalable VLA Training](https://mlsanity.com/paper/2509.22199/) (2025)
- [S-VAM: Shortcut Video-Action Model by Self-Distilling Geometric and Semantic Foresight](https://mlsanity.com/paper/2603.16195/) (2026)
- [Mitty: Diffusion-based Human-to-Robot Video Generation](https://mlsanity.com/paper/2512.17253/) (2025)
- [Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task Generalization](https://mlsanity.com/paper/2608.26103/) (2026)
- [X-Humanoid: Robotize Human Videos to Generate Humanoid Videos at Scale](https://mlsanity.com/paper/2512.04537/) (2025)
- [World Action Models: A Survey](https://mlsanity.com/paper/2606.20781/) (2026)
- [HumanNet: Scaling Human-centric Video Learning to One Million Hours](https://mlsanity.com/paper/2605.06747/) (2026)
- [H2R-Grounder: A Paired-Data-Free Paradigm for Translating Human Interaction Videos into Physically Grounded Robot Videos](https://mlsanity.com/paper/2512.09406/) (2025)
- [HARP-VLA: Human-Robot Aligned Representation Learning for Vision-Language-Action Model](https://mlsanity.com/paper/2605.31234/) (2026)
- [Human-Centric Transferable Tactile Pre-Training for Dexterous Robotic Manipulation](https://mlsanity.com/paper/2607.01067/) (2026)
- [HandEdit: A Unified Benchmark for Egocentric Human-to-Robot Dexterous Hand Image Editing](https://mlsanity.com/paper/2608.12122/) (2026)
- [Ask-to-Clarify: Resolving Instruction Ambiguity through Multi-turn Dialogue](https://mlsanity.com/paper/2509.15061/) (2025)
- [HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments](https://mlsanity.com/paper/2604.14944/) (2026)
- [Trajectory Conditioned Cross-embodiment Skill Transfer](https://mlsanity.com/paper/2510.07773/) (2025)
- [UniHand: A Unified Model for Diverse Controlled 4D Hand Motion Modeling](https://mlsanity.com/paper/2602.21631/) (2026)
- [Hand-Object Interaction in the Age of Large Foundation Models:Reconstruction, Generation, and Embodied Transfer](https://mlsanity.com/paper/2607.28394/) (2026)
- [The Embodiment Gap in Robot Foundation Models](https://mlsanity.com/paper/2608.18433/) (2026)
- [The Imitator Game: Benchmarking Robot Imitative Ability Beyond Action Prediction](https://mlsanity.com/paper/2608.22301/) (2026)
- [GALA: Geometry-Aware Latent Action Modeling for Vision-Language-Action Model Pretraining across Embodiments](https://mlsanity.com/paper/2609.21948/) (2026)
- [Learning Transferable World-Action Models from Task-Paired Human Videos](https://mlsanity.com/paper/2610.14351/) (2026)

## Abstract (arXiv)

Distilling knowledge from human demonstrations is a promising way for robots to learn and act. Existing methods, which often rely on coarsely-aligned video pairs, are typically constrained to learning global or task-level features. As a result, they tend to neglect the fine-grained frame-level dynamics required for complex manipulation and generalization to novel tasks. We posit that this limitation stems from a vicious circle of inadequate datasets and the methods they inspire. To break this cycle, we propose a paradigm shift that treats fine-grained human-robot alignment as a conditional video generation problem. To this end, we first introduce H&R, a novel third-person dataset containing 2,600 episodes of precisely synchronized human and robot motions, collected using a VR teleoperation system. We then present Human2Robot, a framework designed to leverage this data. Human2Robot employs a Video Prediction Model to learn a rich and implicit representation of robot dynamics by generating robot videos from human input, which in turn guides a decoupled action decoder. Our real-world experiments demonstrate that this approach not only achieves high performance on seen tasks but also exhibits significant one-shot generalization to novel positions, objects, instances, and even new task categories.