mlsanityData to 7 Oct 2026

CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models

arXiv 2503.22020 ↗First posted 27 Mar 2025cs.CV cs.AI cs.LGcited by 440

On the map: LLM reasoning → Spatial and embodied reasoning

also in 2 more
Abstract (arXiv)

Vision-language-action models (VLAs) have shown potential in leveraging pretrained vision-language models and diverse robot demonstrations for learning generalizable sensorimotor control. While this paradigm effectively utilizes large-scale data from both robotic and non-robotic sources, current VLAs primarily focus on direct input--output mappings, lacking the intermediate reasoning steps crucial for complex manipulation tasks. As a result, existing VLAs lack temporal planning or reasoning capabilities. In this paper, we introduce a method that incorporates explicit visual chain-of-thought (CoT) reasoning into vision-language-action models (VLAs) by predicting future image frames autoregressively as visual goals before generating a short action sequence to achieve these goals. We introduce CoT-VLA, a state-of-the-art 7B VLA that can understand and generate visual and action tokens. Our experimental results demonstrate that CoT-VLA achieves strong performance, outperforming the state-of-the-art VLA model by 17% in real-world manipulation tasks and 6% in simulation benchmarks. Project website: https://cot-vla.github.io/

Key facts. Cited by 440 papers in the map (data through 7 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 79 of its 84 references that are in the map. First posted on arXiv 27 Mar 2025.

Builds on

79 of 84 references.

Closest ancestors · 6

Same direction, newest first

  1. 2024ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation3 Sep
  2. 2024Robotic Control via Embodied Chain-of-Thought Reasoning11 Jul
  3. 2024Learning Manipulation Skills through Robot Chain-of-Thought with Sparse Failure Guidance22 May
  4. 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities22 Jan
  5. 2023Language to Rewards for Robotic Skill Synthesis14 Jun
  6. 2023EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thought24 May

Other references · 73

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  3. 2023Visual Instruction Tuningcited by 7,792
  4. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  5. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  6. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,243
  7. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  8. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  9. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,037
  10. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  11. 2023Consistency Modelscited by 1,778
  12. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,615
38 more
  1. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,490
  2. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  3. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  4. 2017The "something something" video database for learning and evaluating visual common sensecited by 1,092
  5. 2024Chameleon: Mixed-Modal Early-Fusion Foundation Modelscited by 904
  6. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 895
  7. 2023Open X-Embodiment: Robotic Learning Datasets and RT-X Modelscited by 849
  8. 2024Emu3: Next-Token Prediction is All You Needcited by 761
  9. 2024Show-o: One Single Transformer to Unify Multimodal Understanding and Generationcited by 734
  10. 2022Autoregressive Image Generation using Residual Quantizationcited by 706
  11. 2023BridgeData V2: A Dataset for Robot Learning at Scalecited by 634
  12. 2023GAIA-1: A Generative World Model for Autonomous Drivingcited by 618
  13. 2021CLIPort: What and Where Pathways for Robotic Manipulationcited by 617
  14. 2022ProgPrompt: Generating Situated Robot Task Plans using Large Language Modelscited by 598
  15. 2023Eureka: Human-Level Reward Design via Coding Large Language Modelscited by 542
  16. 2023Learning Universal Policies via Text-Guided Video Generationcited by 506
  17. 2023VideoPoet: A Large Language Model for Zero-Shot Video Generationcited by 479
  18. 2024Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Modelcited by 468
  19. 2024Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generationcited by 447
  20. 2023Learning Interactive Real-World Simulatorscited by 440
  21. 2024Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperationcited by 377
  22. 2022Visual Language Maps for Robot Navigationcited by 362
  23. 2023Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulationcited by 353
  24. 2024Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Modelscited by 336
  25. 2024TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulationcited by 331
  26. 20243D-VLA: A 3D Vision-Language-Action Generative World Modelcited by 320
  27. 2023Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision, Language, Audio, and Actioncited by 301
  28. 2024An Image is Worth 32 Tokens for Reconstruction and Generationcited by 295
  29. 2024VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generationcited by 290
  30. 2024GR-2: A Generative Video-Language-Action Model with Web-Scale Knowledge for Robot Manipulationcited by 290
  31. 2023Zero-Shot Robotic Manipulation with Pretrained Image-Editing Diffusion Modelscited by 264
  32. 2024Visual Sketchpad: Sketching as a Visual Chain of Thought for Multimodal Language Modelscited by 260
  33. 2018Visual Reinforcement Learning with Imagined Goalscited by 248
  34. 2021Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasetscited by 248
  35. 2023Where are we in the search for an Artificial Visual Cortex for Embodied Intelligence?cited by 215
  36. 2024Deep Compression Autoencoder for Efficient High-Resolution Diffusion Modelscited by 209
  37. 2023Any-point Trajectory Modeling for Policy Learningcited by 207
  38. 2018RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitationcited by 202

50 most cited of 73 All 73 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 161; 2026: 279.

440 papers.

Sort:

By area:

  1. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,653
  2. 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningcited by 738
  3. 2026World Action Models are Zero-shot Policiescited by 411
  4. 2026Fast-WAM: Do World Action Models Need Test-time Future Imagination?cited by 311
  5. 2026Causal World Modeling for Robot Controlcited by 289
  6. 2025AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuningcited by 252
  7. 2025DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledgecited by 203
  8. 2025Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontierscited by 193
  9. 2025MolmoAct: Action Reasoning Models that can Reason in Spacecited by 189
  10. 2025MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulationcited by 183
  11. 2025DreamGen: Unlocking Generalization in Robot Learning through Video World Modelscited by 168
  12. 2025LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorizationcited by 166
38 more
  1. 2025FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Drivingcited by 165
  2. 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planningcited by 152
  3. 2025Ctrl-World: A Controllable Generative World Model for Robot Manipulationcited by 149
  4. 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Modelcited by 140
  5. 2025mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAscited by 133
  6. 2025Unified Vision-Language-Action Modelcited by 128
  7. 2026π0.7: a Steerable Generalist Robotic Foundation Model with Emergent Capabilitiescited by 127
  8. 2025Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Modelcited by 125
  9. 2025GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Datacited by 120
  10. 2025NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Taskscited by 119
  11. 2025OneTwoVLA: A Unified Vision-Language-Action Model with Adaptive Reasoningcited by 105
  12. 2025FLARE: Robot Learning with Implicit World Modelingcited by 105
  13. 2026VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Modelcited by 104
  14. 2025A Survey on Vision-Language-Action Models: An Action Tokenization Perspectivecited by 96
  15. 2026ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learningcited by 94
  16. 2025ForceVLA: Enhancing VLA Models with a Force-aware MoE for Contact-rich Manipulationcited by 87
  17. 2025F1: A Vision-Language-Action Model Bridging Understanding and Generation to Actionscited by 85
  18. 2025InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policycited by 83
  19. 2025Vision-Language-Action Models for Robotics: A Review Towards Real-World Applicationscited by 82
  20. 2025Zebra-CoT: A Dataset for Interleaved Vision Language Reasoningcited by 78
  21. 2025Vision-Language-Action (VLA) Models: Concepts, Progress, Applications and Challengescited by 72
  22. 2025InstructVLA: Vision-Language-Action Instruction Tuning from Understanding to Manipulationcited by 71
  23. 2025CogVLA: Cognition-Aligned Vision-Language-Action Model via Instruction-Driven Routing & Sparsificationcited by 70
  24. 2025RynnVLA-002: A Unified Vision-Language-Action and World Modelcited by 70
  25. 2025Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Surveycited by 66
  26. 2025VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulatorscited by 65
  27. 2025WorldEval: World Model as Real-World Robot Policies Evaluatorcited by 64
  28. 2026Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalizationcited by 61
  29. 2025villa-X: Enhancing Latent Action Modeling in Vision-Language-Action Modelscited by 58
  30. 2025RoboMonkey: Scaling Test-Time Sampling and Verification for Vision-Language-Action Modelscited by 56
  31. 2025Training Strategies for Efficient Embodied Reasoningcited by 53
  32. 2025Actions as Language: Fine-Tuning VLMs into VLAs Without Catastrophic Forgettingcited by 52
  33. 2025Interleave-VLA: Enhancing Robot Manipulation with Interleaved Image-Text Instructionscited by 50
  34. 2025FlowVLA: Visual Chain of Thought-based Motion Reasoning for Vision-Language-Action Modelscited by 50
  35. 2025ChatVLA-2: Vision-Language-Action Model with Open-World Embodied Reasoning from Pretrained Knowledgecited by 49
  36. 2026InternVLA-A1: Unifying Understanding, Generation and Action for Robotic Manipulationcited by 48
  37. 2026Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perceptioncited by 47
  38. 2025Pure Vision Language Action (VLA) Models: A Comprehensive Surveycited by 46

50 most cited of 440 All 440 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Vision Language Action Models in Robotic Manipulation: A Systematic Review
  2. 2024Benchmarking Vision, Language, & Action Models on Robotic Learning Tasks
  3. 2026H-VLA: Hierarchical Vision-Language-Action Model with Key-Action Reasoning and Motion Planning in a Unified Action Space
  4. 2024CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation
  5. 2025Do What You Say: Steering Vision-Language-Action Models via Runtime Reasoning-Action Alignment Verification
  6. 2024VLATest: Testing and Evaluating Vision-Language-Action Models for Robotic Manipulation
  7. 2024What Matters in Building Vision-Language-Action Models for Generalist Robots
  8. 2025VLA-0: Building State-of-the-Art VLAs with Zero Modification