mlsanityData to 7 Oct 2026

Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalization

arXiv 2601.12993 ↗First posted 19 Jan 2026cs.ROcited by 61

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

Abstract (arXiv)

We introduce Being-H0.5, a foundational Vision-Language-Action (VLA) model designed for robust cross-embodiment generalization across diverse robotic platforms. While existing VLAs often struggle with morphological heterogeneity and data scarcity, we propose a human-centric learning paradigm that treats human interaction traces as a universal "mother tongue" for physical interaction. To support this, we present UniHand-2.0, the largest embodied pre-training recipe to date, comprising over 35,000 hours of multimodal data across 30 distinct robotic embodiments. Our approach introduces a Unified Action Space that maps heterogeneous robot controls into semantically aligned slots, enabling low-resource robots to bootstrap skills from human data and high-resource platforms. Built upon this human-centric foundation, we design a unified sequential modeling and multi-task pre-training paradigm to bridge human demonstrations and robotic execution. Architecturally, Being-H0.5 utilizes a Mixture-of-Transformers design featuring a novel Mixture-of-Flow (MoF) framework to decouple shared motor primitives from specialized embodiment-specific experts. Finally, to make cross-embodiment policies stable in the real world, we introduce Manifold-Preserving Gating for robustness under sensory shift and Universal Async Chunking to universalize chunked control across embodiments with different latency and control profiles. We empirically demonstrate that Being-H0.5 achieves state-of-the-art results on simulated benchmarks, such as LIBERO (98.9%) and RoboCasa (53.9%), while also exhibiting strong cross-embodiment capabilities on five robotic platforms.

Key facts. Cited by 61 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 121 of its 133 references that are in the map. First posted on arXiv 19 Jan 2026.

Builds on

121 of 133 references.

Closest ancestors · 7

Same direction, newest first

  1. 2025METIS: Multi-Source Egocentric Training for Integrated Dexterous Vision-Language-Action Model21 Nov
  2. 2025Scalable Vision-Language-Action Model Pretraining for Robotic Manipulation with Real-Life Human Activity Videos24 Oct
  3. 2025Being-H0: Vision-Language-Action Pretraining from Large-Scale Human Videos21 Jul
  4. 2025EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos16 Jul
  5. 2025UniVLA: Learning to Act Anywhere with Task-centric Latent Actions9 May
  6. 2025DexGraspVLA: A Vision-Language-Action Framework Towards General Dexterous Grasping28 Feb
  7. 2024Latent Action Pretraining from Videos15 Oct

Other references · 114

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  5. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  6. 2022Flow Matching for Generative Modelingcited by 5,426
  7. 2025Qwen2.5-VL Technical Reportcited by 5,418
  8. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  9. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,151
  10. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  11. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  12. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
38 more
  1. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  2. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  3. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  4. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  5. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  6. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  7. 2018SlowFast Networks for Video Recognitioncited by 1,862
  8. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,653
  9. 2021Ego4D: Around the World in 3,000 Hours of Egocentric Videocited by 1,530
  10. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,419
  11. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  12. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  13. 2019Cross-lingual Language Model Pretrainingcited by 1,088
  14. 2016Modeling Context in Referring Expressionscited by 1,054
  15. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,023
  16. 2024Grounded SAM: Assembling Open-World Models for Diverse Visual Taskscited by 914
  17. 2025Emerging Properties in Unified Multimodal Pretrainingcited by 855
  18. 2023Open X-Embodiment: Robotic Learning Datasets and RT-X Modelscited by 849
  19. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 844
  20. 2024DROID: A Large-Scale In-The-Wild Robot Manipulation Datasetcited by 801
  21. 2018Scaling Egocentric Vision: The EPIC-KITCHENS Datasetcited by 789
  22. 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 745
  23. 2022R3M: A Universal Visual Representation for Robot Manipulationcited by 678
  24. 2024RDT-1B: a Diffusion Foundation Model for Bimanual Manipulationcited by 670
  25. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 643
  26. 2023BridgeData V2: A Dataset for Robot Learning at Scalecited by 634
  27. 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Roboticscited by 507
  28. 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Modelcited by 502
  29. 2023Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectivescited by 477
  30. 2025CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Modelscited by 440
  31. 2024CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulationcited by 411
  32. 2024RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robotscited by 391
  33. 2025AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systemscited by 390
  34. 2022VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Trainingcited by 384
  35. 2024Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robotscited by 373
  36. 2023Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulationcited by 353
  37. 2024Robotic Control via Embodied Chain-of-Thought Reasoningcited by 340
  38. 2022HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object Interactioncited by 321

50 most cited of 114 All 114 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 61.

61 papers.

Sort:

By area:

  1. 2026π0.7: a Steerable Generalist Robotic Foundation Model with Emergent Capabilitiescited by 127
  2. 2026Being-H0.7: A Latent World-Action Model from Egocentric Videoscited by 68
  3. 2026Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Modelscited by 47
  4. 2026World Model for Robot Learning: A Comprehensive Surveycited by 42
  5. 2026Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodimentscited by 42
  6. 2026ABot-M0.5: Unified Mobility-and-Manipulation World Action Modelcited by 28
  7. 2026From Foundation to Application: Improving VLA Models in Practicecited by 25
  8. 2026LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfercited by 17
  9. 2026UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Drivingcited by 17
  10. 2026G0.5: One Autoregressive Stream for Robot Reasoning and Actioncited by 17
  11. 2026Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Modelcited by 13
  12. 2026chi0: Resource-Aware Robust Manipulation via Taming Distributional Inconsistenciescited by 11
38 more
  1. 2026Egocentric World Model for Photorealistic Hand-Object Interaction Synthesiscited by 10
  2. 2026HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulationcited by 10
  3. 2026HumanNet: Scaling Human-centric Video Learning to One Million Hourscited by 9
  4. 2026DeMaVLA: A Vision-Language-Action Foundation Model for Generalizable Deformable Manipulationcited by 8
  5. 2026Kairos: A Regret-Aware Native World-Action Model Stack for Physical AIcited by 8
  6. 2026Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularizationcited by 7
  7. 2026PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interactioncited by 6
  8. 2026DexHiL: A Human-in-the-Loop Framework for Vision-Language-Action Model Post-Training in Dexterous Manipulationcited by 5
  9. 2026Human Cognition in Machines: A Unified Perspective of World Modelscited by 5
  10. 2026Open-AoE: An Open Egocentric Manipulation Dataset and Toolchain for Embodied Learningcited by 5
  11. 2026ActiveMimic: Egocentric Video Pretraining with Active Perceptioncited by 4
  12. 2026Human-Centric Transferable Tactile Pre-Training for Dexterous Robotic Manipulationcited by 4
  13. 2026Data Pyramid for Embodied Manipulation: A Surveycited by 4
  14. 2026DECO: Decoupled Multimodal Diffusion Transformer for Bimanual Dexterous Manipulation with a Plugin Tactile Adaptercited by 3
  15. 2026ArtHOI: Taming Foundation Models for Monocular 4D Reconstruction of Hand-Articulated-Object Interactionscited by 3
  16. 2026GaussianDream: A Feed-Forward 3D Gaussian World Model for Robotic Manipulationcited by 3
  17. 2026τ0-VLA: a Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Computationcited by 3
  18. 2026NymeriaPlus: Enriching Nymeria Dataset with Additional Annotations and Datacited by 2
  19. 2026QuoVLA: Quotient Space for Vision-Language-Action Modelscited by 2
  20. 2026ω-EVA: Envision, Verify, and Act with Latent Interactive World Modelscited by 2
  21. 2026HumanScale: Egocentric Human Video Can Outperform Real-Robot Data for Embodied Pretrainingcited by 2
  22. 2026Wh0: Generative World Models as Scalable Sources of Egocentric Human Hand Manipulation Datacited by 2
  23. 2026Seeing Touch from Motion: A Unified Modality-Aware Visuo-Tactile Policy with Tactile Motion Correlationcited by 2
  24. 2026LD4WAM: Learning Latent Dynamics from Human Videos for World Action Modelscited by 2
  25. 2026Riemann-1.0: An Embodied World Action Model for Physical AIcited by 2
  26. 2026Unmasking the Illusion of Embodied Reasoning in Vision-Language-Action Modelscited by 1
  27. 2026BORA: Bridging Offline Reinforcement Learning and Online Residual Adaptation for Real-World Dexterous VLA Modelscited by 1
  28. 2026Translation as a Bridging Action: Transferring Manipulation Skills from Humans to Robotscited by 1
  29. 2026Hermite Curves as Trajectory Priors for Vision-Language-Action Modelscited by 1
  30. 2026Any2Any: Efficient Cross-Embodiment Transfer for Humanoid Whole-Body Trackingcited by 0
  31. 2026Unified Video-Action Joint Denoising for Dexterous Action and Data Generationcited by 0
  32. 2026DexPIE: Stable Dexterous Policy Improvement from Real-World Experiencecited by 0
  33. 2026V2P-Manip: Learning Dexterous Manipulation from Monocular Human Videoscited by 0
  34. 2026EgoRecovery: Acquiring Failure Recovery Ability Through Human Recovery Demonstrationcited by 0
  35. 2026XEWorld: Can Action-Conditioned World Models Generalize to Unseen Robot Embodiments?cited by 0
  36. 2026One Policy, Many Embodiments: Unified Camera-Centric Action Geometry Pre-training for Heterogeneous Embodied Manipulationcited by 0
  37. 2026ZETA: A Controlled Study of Zero-Shot Cross-Embodiment VLA Transfer for Tabletop Manipulationcited by 0
  38. 2026DeCAL: Towards Physically-Grounded Dexterous Vision-Language-Action Models via Contact-Aware Latent Co-Imaginationcited by 0

50 most cited of 61 All 61 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Emergence of Human to Robot Transfer in Vision-Language-Action Models
  2. 2026GigaBrain-0.7: Scaling Embodied Foundation Models to Emergent Capabilities with a Three-System Architecture
  3. 2026RDT2: Exploring the Scaling Limit of UMI Data Towards Zero-Shot Cross-Embodiment Generalization
  4. 2026DyPES-VLA: Learning Shared Dynamics Priors and Embodiment-Specific Control for Cross-Embodiment Manipulation
  5. 2026JoyAI-RA 0.1: A Foundation Model for Robotic Autonomy
  6. 2025Embodiment Transfer Learning for Vision-Language-Action Models
  7. 2026Vision-Language-Action in Robotics: A Survey of Datasets, Benchmarks, and Data Engines
  8. 2026From Human Videos to Robot Manipulation: A Survey on Scalable Vision-Language-Action Learning with Human-Centric Data