mlsanityData to 7 Oct 2026

WorldVLA: Towards Autoregressive Action World Model

arXiv 2506.21539 ↗First posted 26 Jun 2025cs.RO cs.AIcited by 351

On the map: World models → Embodied and robot world models

Abstract (arXiv)

We present WorldVLA, an autoregressive action world model that unifies action and image understanding and generation. Our WorldVLA intergrates Vision-Language-Action (VLA) model and world model in one single framework. The world model predicts future images by leveraging both action and image understanding, with the purpose of learning the underlying physics of the environment to improve action generation. Meanwhile, the action model generates the subsequent actions based on image observations, aiding in visual understanding and in turn helps visual generation of the world model. We demonstrate that WorldVLA outperforms standalone action and world models, highlighting the mutual enhancement between the world model and the action model. In addition, we find that the performance of the action model deteriorates when generating sequences of actions in an autoregressive manner. This phenomenon can be attributed to the model's limited generalization capability for action prediction, leading to the propagation of errors from earlier actions to subsequent ones. To address this issue, we propose an attention mask strategy that selectively masks prior actions during the generation of the current action, which shows significant performance improvement in the action chunk generation task.

Key facts. Cited by 351 papers in the map (data through 7 Oct 2026). Placed in Embodied and robot world models, one of 11 directions in World models. Builds on 46 of its 50 references that are in the map. First posted on arXiv 26 Jun 2025.

Builds on

46 of 50 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulation19 Dec
  2. 2024iVideoGPT: Interactive VideoGPTs are Scalable World Models24 May
  3. 20243D-VLA: A 3D Vision-Language-Action Generative World Model14 Mar
  4. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar

Other references · 42

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2023Visual Instruction Tuningcited by 7,792
  5. 2025Qwen2.5-VL Technical Reportcited by 5,418
  6. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  7. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,048
  8. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  9. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,898
  10. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  11. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  12. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,526
30 more
  1. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  2. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  3. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,653
  4. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,490
  5. 2018World Modelscited by 1,444
  6. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  7. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  8. 2024Chameleon: Mixed-Modal Early-Fusion Foundation Modelscited by 904
  9. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 844
  10. 2025Cosmos World Foundation Model Platform for Physical AIcited by 833
  11. 2025Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scalingcited by 787
  12. 2024Emu3: Next-Token Prediction is All You Needcited by 761
  13. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 643
  14. 2023Learning Universal Policies via Text-Guided Video Generationcited by 506
  15. 2025VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understandingcited by 502
  16. 2022Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priorscited by 487
  17. 2024Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Modelcited by 468
  18. 2023An Embodied Generalist Agent in 3D Worldcited by 355
  19. 2023Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulationcited by 353
  20. 2023Vision-Language Foundation Models as Effective Robot Imitatorscited by 344
  21. 2022MAGVIT: Masked Generative Video Transformercited by 341
  22. 2024TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulationcited by 331
  23. 2024GR-2: A Generative Video-Language-Action Model with Web-Scale Knowledge for Robot Manipulationcited by 290
  24. 2024MetaMorph: Multimodal Understanding and Generation via Instruction Tuningcited by 205
  25. 2025Unified Video Action Modelcited by 195
  26. 2023Compositional Foundation Models for Hierarchical Planningcited by 92
  27. 2024Closed-Loop Visuomotor Control with Generative Expectation for Robotic Manipulationcited by 51
  28. 2025Pre-Trained Video Generative Models as World Simulatorscited by 43
  29. 2024Diffusion Transformer Policycited by 41
  30. 2024Using Left and Right Brains Together: Towards Vision and Language Planningcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 64; 2026: 287.

351 papers.

Sort:

By area:

  1. 2026Fast-WAM: Do World Action Models Need Test-time Future Imagination?cited by 311
  2. 2025LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Modelscited by 278
  3. 2025MolmoAct: Action Reasoning Models that can Reason in Spacecited by 189
  4. 2025LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorizationcited by 166
  5. 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Modelcited by 140
  6. 2026GigaWorld-Policy: An Efficient Action-Centered World--Action Modelcited by 129
  7. 2025DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Drivingcited by 127
  8. 2026VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Modelcited by 104
  9. 2026ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learningcited by 94
  10. 2025InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policycited by 83
  11. 2025Vision-Language-Action Models for Robotics: A Review Towards Real-World Applicationscited by 82
  12. 2025ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceivercited by 80
38 more
  1. 2026DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Controlcited by 71
  2. 2025RynnVLA-002: A Unified Vision-Language-Action and World Modelcited by 70
  3. 2025Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Surveycited by 66
  4. 2025VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulatorscited by 65
  5. 2025A Comprehensive Survey on World Models for Embodied AIcited by 65
  6. 2026Unified 4D World Action Modeling from Video Priors with Asynchronous Denoisingcited by 65
  7. 2025FlowVLA: Visual Chain of Thought-based Motion Reasoning for Vision-Language-Action Modelscited by 50
  8. 2026InternVLA-A1: Unifying Understanding, Generation and Action for Robotic Manipulationcited by 48
  9. 2025Pure Vision Language Action (VLA) Models: A Comprehensive Surveycited by 46
  10. 2026World Action Models: The Next Frontier in Embodied AIcited by 46
  11. 2025Unified Diffusion VLA: Vision-Language-Action Model via Joint Discrete Denoising Diffusion Processcited by 37
  12. 2026Do World Action Models Generalize Better than VLAs? A Robustness Studycited by 35
  13. 2025The Better You Learn, The Smarter You Prune: Towards Efficient Vision-language-action Models via Differentiable Token Pruningcited by 34
  14. 2025Dual-Stream Diffusion for World-Model Augmented Vision-Language-Action Modelcited by 33
  15. 2026ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?cited by 33
  16. 2025Action-aware Dynamic Pruning for Efficient Vision-Language-Action Manipulationcited by 32
  17. 2025VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layercited by 32
  18. 2025Vision-Language-Action Models for Autonomous Driving: Past, Present, and Futurecited by 32
  19. 2026World Guidance: World Modeling in Condition Space for Action Generationcited by 32
  20. 2026MIND: Benchmarking Memory Consistency and Action Control in World Modelscited by 31
  21. 2025dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thoughtcited by 30
  22. 2025PAI-Bench: A Comprehensive Benchmark For Physical AIcited by 30
  23. 2025Astra: General Interactive World Model with Autoregressive Denoisingcited by 30
  24. 2026ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Modelscited by 30
  25. 2026Native Video-Action Pretraining for Generalizable Robot Controlcited by 30
  26. 2026LaST0: Latent Spatio-Temporal Chain-of-Thought for Robotic Vision-Language-Action Modelcited by 28
  27. 2026ABot-M0.5: Unified Mobility-and-Manipulation World Action Modelcited by 28
  28. 2026When Vision Overrides Language: Evaluating and Mitigating Counterfactual Failures in VLAscited by 27
  29. 2026StarVLA-α: Reducing Complexity in Vision-Language-Action Systemscited by 27
  30. 2025NORA-1.5: A Vision-Language-Action Model Trained using World Model- and Action-based Preference Rewardscited by 26
  31. 2025SRPO: Self-Referential Policy Optimization for Vision-Language-Action Modelscited by 25
  32. 2026OA-WAM: Object-Addressable World Action Model for Robust Robot Manipulationcited by 25
  33. 2025RynnVLA-001: Using Human Demonstrations to Improve Robot Manipulationcited by 24
  34. 2026Adaptive Action Chunking at Inference-time for Vision-Language-Action Modelscited by 24
  35. 2026WorldMark: A Unified Benchmark Suite for Interactive Video World Modelscited by 24
  36. 2026GigaBrain-0.5M*: a VLA That Learns From World Model-Based Reinforcement Learningcited by 23
  37. 2025MLA: A Multisensory Language-Action Model for Multimodal Understanding and Forecasting in Robotic Manipulationcited by 22
  38. 2025AVA-VLA: Improving Vision-Language-Action models with Active Visual Attentioncited by 22

50 most cited of 351 All 351 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Devol-ONE: One Autoregressive Mixture of Transformers to Unify Vision-Language-Action and Latent World Modeling
  2. 2026World-Language-Action Model for Unified World Modeling, Language Reasoning, and Action Synthesis
  3. 2026UniWAM: Unified World-Action Model
  4. 2026World2Act: Latent Action Post-Training from World Model Dynamics
  5. 2026WorldAgen: Unified State-Action Prediction with Test-Time World Model Training
  6. 2026V-JEPA Policy: Building Effective World-Action Models on Predictive Visual Latents
  7. 2026World Pilot: Steering Vision-Language-Action Models with World-Action Priors
  8. 2026World-Calibrated Proposal-to-Action Flow for Vision-Language-Action Models