mlsanityData to 7 Oct 2026

Emergence of Human to Robot Transfer in Vision-Language-Action Models

arXiv 2512.22414 ↗First posted 27 Dec 2025cs.RO cs.AIcited by 36

On the map: Vision-language-action models → Human-video and latent-action pretraining

Abstract (arXiv)

Vision-language-action (VLA) models can enable broad open world generalization, but require large and diverse datasets. It is appealing to consider whether some of this data can come from human videos, which cover diverse real-world situations and are easy to obtain. However, it is difficult to train VLAs with human videos alone, and establishing a mapping between humans and robots requires manual engineering and presents a major research challenge. Drawing inspiration from advances in large language models, where the ability to learn from diverse supervision emerges with scale, we ask whether a similar phenomenon holds for VLAs that incorporate human video data. We introduce a simple co-training recipe, and find that human-to-robot transfer emerges once the VLA is pre-trained on sufficient scenes, tasks, and embodiments. Our analysis suggests that this emergent capability arises because diverse pretraining produces embodiment-agnostic representations for human and robot data. We validate these findings through a series of experiments probing human to robot skill transfer and find that with sufficiently diverse robot pre-training our method can nearly double the performance on generalization settings seen only in human data.

Key facts. Cited by 36 papers in the map (data through 7 Oct 2026). Placed in Human-video and latent-action pretraining, one of 13 directions in Vision-language-action models. Builds on 54 of its 69 references that are in the map. First posted on arXiv 27 Dec 2025.

Builds on

54 of 69 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs17 Dec
  2. 2025EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos16 Jul
  3. 2024Latent Action Pretraining from Videos15 Oct

Other references · 51

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  3. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  4. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  5. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  6. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,037
  7. 2021Ego4D: Around the World in 3,000 Hours of Egocentric Videocited by 1,530
  8. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  9. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  10. 2024DROID: A Large-Scale In-The-Wild Robot Manipulation Datasetcited by 801
  11. 2022R3M: A Universal Visual Representation for Robot Manipulationcited by 678
  12. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 643
38 more
  1. 2023BridgeData V2: A Dataset for Robot Learning at Scalecited by 634
  2. 2025AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systemscited by 390
  3. 2022VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Trainingcited by 384
  4. 2024Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robotscited by 373
  5. 2024TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulationcited by 331
  6. 20243D-VLA: A 3D Vision-Language-Action Generative World Modelcited by 320
  7. 2025X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Modelcited by 285
  8. 2019RoboNet: Large-Scale Multi-Robot Learningcited by 268
  9. 2021H2O: Two Hands Manipulating Objects for First Person Interaction Recognitioncited by 251
  10. 2021Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasetscited by 248
  11. 2023Affordances from Human Videos as a Versatile Representation for Roboticscited by 221
  12. 2023Where are we in the search for an Artificial Visual Cortex for Embodied Intelligence?cited by 215
  13. 2023MimicPlay: Long-Horizon Imitation Learning by Watching Human Playcited by 214
  14. 2023Project Aria: A New Tool for Egocentric Multi-Modal AI Researchcited by 201
  15. 2025DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Controlcited by 191
  16. 2023RH20T: A Comprehensive Robotic Dataset for Learning Diverse Skills in One-Shotcited by 173
  17. 2023RoboAgent: Generalization and Efficiency in Robot Manipulation via Semantic Augmentations and Action Chunkingcited by 167
  18. 2024EgoMimic: Scaling Imitation Learning via Egocentric Videocited by 160
  19. 2024Behavior Generation with Latent Actionscited by 159
  20. 2022Human-to-Robot Imitation in the Wildcited by 157
  21. 2024DexCap: Scalable and Portable Mocap Data Collection System for Dexterous Manipulationcited by 149
  22. 2025Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Bettercited by 142
  23. 2021Learning Generalizable Robotic Reward Functions from "In-The-Wild" Human Videoscited by 116
  24. 2024Track2Act: Predicting Point Tracks from Internet Videos enables Generalizable Robot Manipulationcited by 116
  25. 2020Visual Imitation Made Easycited by 111
  26. 2024Scaling Cross-Embodied Learning: One Policy for Manipulation, Navigation, Locomotion and Aviationcited by 108
  27. 2019Grasping in the Wild:Learning 6DoF Closed-Loop Grasping from Low-Cost Demonstrationscited by 104
  28. 2023On Bringing Robots Homecited by 101
  29. 2025Galaxea Open-World Dataset and G0 Dual-System VLA Modelcited by 90
  30. 2025Phantom: Training Robots Without Robots Using Only Human Videoscited by 64
  31. 2024Pushing the Limits of Cross-Embodiment Learning for Manipulation and Navigationcited by 60
  32. 2025DexUMI: Using Human Hand as the Universal Manipulation Interface for Dexterous Manipulationcited by 58
  33. 2024Dyn-HaMR: Recovering 4D Interacting Hand Motion from a Dynamic Cameracited by 50
  34. 2024From Multimodal LLMs to Generalist Embodied Agents: Methods and Lessonscited by 42
  35. 2025DexWild: Dexterous Human Interactions for In-the-Wild Robot Policiescited by 34
  36. 2025EgoZero: Robot Learning from Smart Glassescited by 28
  37. 2025EgoBridge: Domain Adaptation for Generalizable Imitation from Egocentric Human Datacited by 27
  38. 2025EMMA: Scaling Mobile Manipulation via Egocentric Human Datacited by 20

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 36.

36 papers.

Sort:

By area:

  1. 2026World Action Models are Zero-shot Policiescited by 411
  2. 2026π0.7: a Steerable Generalist Robotic Foundation Model with Emergent Capabilitiescited by 127
  3. 2026DreamDojo: A Generalist Robot World Model from Large-Scale Human Videoscited by 117
  4. 2026EgoVerse: An Egocentric Human Dataset for Robot Learning from Around the Worldcited by 49
  5. 2026InternVLA-A1: Unifying Understanding, Generation and Action for Robotic Manipulationcited by 48
  6. 2026FRAPPE: Infusing World Modeling into Generalist Policies via Multiple Future Representation Alignmentcited by 19
  7. 2026EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Datacited by 17
  8. 2026RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulationcited by 15
  9. 2026HumanEgo: Zero-Shot Robot Learning from Minutes of Human Egocentric Videoscited by 14
  10. 2026VLANeXt: Recipes for Building Strong VLA Modelscited by 12
  11. 2026T-Rex: Tactile-Reactive Dexterous Manipulationcited by 10
  12. 2026UniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modelingcited by 9
24 more
  1. 2026ActiveMimic: Egocentric Video Pretraining with Active Perceptioncited by 4
  2. 2026A Mechanistic Analysis of Sim-and-Real Co-Training in Generative Robot Policiescited by 3
  3. 2026From Human Videos to Robot Manipulation: A Survey on Scalable Vision-Language-Action Learning with Human-Centric Datacited by 3
  4. 2026From Pixels to Tokens: A Systematic Study of Latent Action Supervision for Vision-Language-Action Modelscited by 2
  5. 2026Ego-Pi: VLA Fine-Tuning for Ego-Centric Human and Robot Datacited by 2
  6. 2026Wh0: Generative World Models as Scalable Sources of Egocentric Human Hand Manipulation Datacited by 2
  7. 2026Towards Predictive, Aligned, and Scalable Robot Learningcited by 2
  8. 2026EgoTac: In-the-wild Tactile Prediction from Egocentric Visioncited by 2
  9. 2026What Matters for Latent Actions in Robot Learningcited by 2
  10. 2026LD4WAM: Learning Latent Dynamics from Human Videos for World Action Modelscited by 2
  11. 2026LIDEA: Human-to-Robot Imitation Learning via Implicit Feature Distillation and Explicit Geometry Alignmentcited by 1
  12. 2026RIO: Flexible Real-Time Robot I/O for Cross-Embodiment Robot Learningcited by 1
  13. 2026Translation as a Bridging Action: Transferring Manipulation Skills from Humans to Robotscited by 1
  14. 2026Hand-Object Interaction in the Age of Large Foundation Models:Reconstruction, Generation, and Embodied Transfercited by 0
  15. 2026ACE-Data-0: Human-Centric Ambient Capture as Embodied Data Enginecited by 0
  16. 2026BRIDGE: An Open-Source Humanoid Platform via Morphology-Control Co-Design for Physical AIcited by 0
  17. 2026ReWeight: Leveraging Human Data for VLA Post-Training via Demonstration Retrieval and Sample Weightingcited by 0
  18. 2026ActiveScale: Scaling Active Perception for Robots across Model, Data, and Hardwarecited by 0
  19. 2026Improving Cross-embodiment Transfer in Latent Action Models with Action-Similarity Supervisioncited by 0
  20. 2026Skel-WAM: A Hand-Skeleton-Conditioned World Action Model for Human-to-Robot Manipulation Transfercited by 0
  21. 2026Intelligence Across Embodimentscited by 0
  22. 2026EgoSpeedUp: Transferring Human Manipulation Tempo to Robot Policiescited by 0
  23. 2026EgoHumanoid-V2: Human-to-Humanoid Transfer of Coordinated Whole-Body Skills for Loco-Manipulationcited by 0
  24. 2026Ego4WAM: What Matters When Scaling Egocentric Human Data for Robot Learning?cited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization
  2. 2025From Intention to Execution: Probing the Generalization Boundaries of Vision-Language-Action Models
  3. 2026Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalization
  4. 2025MiVLA: Towards Generalizable Vision-Language-Action Model with Human-Robot Mutual Imitation Pre-training
  5. 2026Vision-Language-Action in Robotics: A Survey of Datasets, Benchmarks, and Data Engines
  6. 2026Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data
  7. 2025Embodiment Transfer Learning for Vision-Language-Action Models
  8. 2025GigaBrain-0: A World Model-Powered Vision-Language-Action Model