mlsanityData to 7 Oct 2026

VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulators

arXiv 2510.00406 ↗First posted 1 Oct 2025cs.RO cs.CVcited by 65

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

also in 2 more
Abstract (arXiv)

Vision-Language-Action (VLA) models enable embodied decision-making but rely heavily on imitation learning, leading to compounding errors and poor robustness under distribution shift. Reinforcement learning (RL) can mitigate these issues yet typically demands costly real-world interactions or suffers from sim-to-real gaps. We introduce VLA-RFT, a reinforcement fine-tuning framework that leverages a data-driven world model as a controllable simulator. Trained from real interaction data, the simulator predicts future visual observations conditioned on actions, allowing policy rollouts with dense, trajectory-level rewards derived from goal-achieving references. This design delivers an efficient and action-aligned learning signal, drastically lowering sample requirements. With fewer than 400 fine-tuning steps, VLA-RFT surpasses strong supervised baselines and achieves greater efficiency than simulator-based RL. Moreover, it exhibits strong robustness under perturbed conditions, sustaining stable task execution. Our results establish world-model-based RFT as a practical post-training paradigm to enhance the generalization and robustness of VLA models. For more details, please refer to https://vla-rft.github.io/.

Key facts. Cited by 65 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 63 of its 69 references that are in the map. First posted on arXiv 1 Oct 2025.

Builds on

63 of 69 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025FlowVLA: Visual Chain of Thought-based Motion Reasoning for Vision-Language-Action Models25 Aug
  2. 2025WorldVLA: Towards Autoregressive Action World Model26 Jun
  3. 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning11 Jun
  4. 2024iVideoGPT: Interactive VideoGPTs are Scalable World Models24 May
  5. 2024RoboDreamer: Learning Compositional World Models for Robot Imagination18 Apr
  6. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar

Other references · 57

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  4. 2022Scalable Diffusion Models with Transformerscited by 5,931
  5. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  6. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  7. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  8. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,526
  9. 2022Video Diffusion Modelscited by 2,280
  10. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  11. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 1,980
  12. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,598
38 more
  1. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  2. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  3. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,221
  4. 2023Mastering Diverse Domains through World Modelscited by 1,193
  5. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  6. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 844
  7. 2021CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Taskscited by 564
  8. 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Modelcited by 502
  9. 2024Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Modelscited by 467
  10. 2025CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Modelscited by 440
  11. 2022DayDreamer: World Models for Physical Robot Learningcited by 372
  12. 2024DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planningcited by 357
  13. 2024Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Modelscited by 336
  14. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 298
  15. 2019Causal Confusion in Imitation Learningcited by 286
  16. 2024Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representationscited by 279
  17. 2025VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learningcited by 166
  18. 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Modelcited by 140
  19. 2025ConRFT: A Reinforced Fine-tuning Method for VLA Models via Consistency Policycited by 139
  20. 2024Multimodal Diffusion Transformer: Learning Versatile Behavior from Multimodal Goalscited by 131
  21. 2025Interactive Post-Training for Vision-Language-Action Modelscited by 129
  22. 2025Improving Vision-Language-Action Model with Online Reinforcement Learningcited by 107
  23. 2024GRAPE: Generalizing Robot Policy via Preference Alignmentcited by 106
  24. 2025What Can RL Bring to VLA Generalization? An Empirical Studycited by 102
  25. 2025ReinFlow: Fine-tuning Flow Matching Policy with Online Reinforcement Learningcited by 83
  26. 2025OpenHelix: A Short Survey, Empirical Analysis, and Open-Source Dual-System VLA Model for Robotic Manipulationcited by 79
  27. 2024Is Behavior Cloning All You Need? Understanding Horizon in Imitation Learningcited by 74
  28. 2024Policy Agnostic RL: Offline RL and Online RL Fine-Tuning of Any Class and Backbonecited by 71
  29. 2025Eagle 2: Building Post-Training Data Strategies from Scratch for Frontier Vision-Language Modelscited by 69
  30. 2025A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learningcited by 66
  31. 2023QUAR-VLA: Vision-Language-Action Model for Quadruped Robotscited by 55
  32. 2025Long-VLA: Unleashing Long-Horizon Capability of Vision Language Action Model for Robot Manipulationcited by 48
  33. 2025Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Roboticscited by 47
  34. 2025TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimizationcited by 47
  35. 2025ReinboT: Amplifying Robot Visual-Language Manipulation with Reinforcement Learningcited by 44
  36. 2025Humanoid-VLA: Towards Universal Humanoid Control with Visual Integrationcited by 43
  37. 2024RLDG: Robotic Generalist Policy Distillation via Reinforcement Learningcited by 42
  38. 2022Curious Exploration via Structured World Models Yields Zero-Shot Object Manipulationcited by 29

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 61.

65 papers.

Sort:

By area:

  1. 2026GigaWorld-Policy: An Efficient Action-Centered World--Action Modelcited by 129
  2. 2026DreamDojo: A Generalist Robot World Model from Large-Scale Human Videoscited by 117
  3. 2026WoVR: World Models as Reliable Simulators for Post-Training VLA Policies with RLcited by 46
  4. 2026World Action Models: The Next Frontier in Embodied AIcited by 46
  5. 2026World Model for Robot Learning: A Comprehensive Surveycited by 42
  6. 2026VLAW: Iterative Co-Improvement of Vision-Language-Action Policy and World Modelcited by 39
  7. 2026World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policycited by 36
  8. 2025A Survey on Efficient Vision-Language-Action Modelscited by 34
  9. 2026ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Modelscited by 30
  10. 2026AtomVLA: Scalable Post-Training for Robotic Manipulation via Predictive Latent World Modelscited by 21
  11. 2025TiViBench: Benchmarking Think-in-Video Reasoning for Video Generative Modelscited by 18
  12. 2026On-the-Fly VLA Adaptation via Test-Time Reinforcement Learningcited by 14
38 more
  1. 2026World-Value-Action Model: Implicit Planning for Vision-Language-Action Systemscited by 13
  2. 2025Affordance Field Intervention: Enabling VLAs to Escape Memory Traps in Robotic Manipulationcited by 12
  3. 2025EvoVLA: Self-Evolving Vision-Language-Action Modelcited by 11
  4. 2026Wow, wo, val! A Comprehensive Embodied World Model Evaluation Turing Testcited by 11
  5. 2026Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuningcited by 11
  6. 2026RoboStereo: Dual-Tower 4D Embodied World Models for Unified Policy Optimizationcited by 9
  7. 2026VLA-OPD: Bridging Offline SFT and Online RL for Vision-Language-Action Models via On-Policy Distillationcited by 9
  8. 2026How Should World Models Be Evaluated for Embodied Decision-Making? A Decision-Making-Centric Positioncited by 9
  9. 2026ReWorld: Multi-Dimensional Reward Modeling for Embodied World Modelscited by 7
  10. 2026RL-VLA3: A Flexible and Asynchronous Reinforcement Learning Framework for VLA Trainingcited by 7
  11. 2026World2Act: Latent Action Post-Training from World Model Dynamicscited by 7
  12. 2026OmniGuide: Universal Guidance Fields for Enhancing Generalist Robot Policiescited by 6
  13. 2026Orca: The World is in Your Mindcited by 6
  14. 2026VLA-Thinker: Boosting Vision-Language-Action Models through Thinking-with-Image Reasoningcited by 5
  15. 2026PhyAgentOS: A Self-Evolving Operating System for Embodied Agents with Decoupled Cognitive Planning and Physical Executioncited by 5
  16. 2026Self-Correcting VLA: Online Action Refinement via Sparse World Imaginationcited by 4
  17. 2026Towards Practical World Model-based Reinforcement Learning for Vision-Language-Action Modelscited by 4
  18. 2026Data Pyramid for Embodied Manipulation: A Surveycited by 4
  19. 2026What Makes Value Learning Efficient in Residual Reinforcement Learning?cited by 3
  20. 2026Learning Action Manifold with Multi-view Latent Priors for Robotic Manipulationcited by 3
  21. 2026WAM-RL: World-Action Model Reinforcement Learning with Reconstruction Rewards and Online Video SFTcited by 3
  22. 2026Learning Transferable Dynamics Priors from Action to World Modelingcited by 3
  23. 2026FACT: Failure-Aware Causal Training for World-Action Modelscited by 3
  24. 2026Scaling Sim-to-Real VLA Reinforcement Learning with Generative 3D Worldscited by 2
  25. 2026World Pilot: Steering Vision-Language-Action Models with World-Action Priorscited by 2
  26. 2026PearlVLA: Progressive Embodied Action-Plan Refinement in Latent Spacecited by 2
  27. 2026SafeDojo: Safe Reinforcement Learning for VLA via Interactive World Modelcited by 2
  28. 2026WorldSample: Closed-loop Real-robot RL with World Modellingcited by 2
  29. 2026ExToken: Structured Exploration for Efficient Vision-Language-Action Reinforcement Fine-tuningcited by 2
  30. 2026PRM-as-a-Judge 1.5: A Toolkit for Robot Process Assessmentcited by 2
  31. 2026NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Modelscited by 1
  32. 2026Reinforcing Structured Chain-of-Thought for Video Understandingcited by 1
  33. 2026CounterAlign: Counterfactual Supervision for Vision-Language-Action Modelscited by 1
  34. 2026World Models for Embodied Intelligence: From Plausible to Controllable to Actionablecited by 1
  35. 2026Navigating the Clutter: Waypoint-Based Bi-Level Planning for Multi-Robot Systemscited by 0
  36. 2026Reinforcing VLAs in Task-Agnostic World Modelscited by 0
  37. 2026Learn Where Outcomes Diverge: Efficient VLA RL via Probabilistic Chunk Maskingcited by 0
  38. 2026Foresight: Iterative Reasoning About Clues that Matter for Navigationcited by 0

50 most cited of 65 All 65 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025CO-RFT: Efficient Fine-Tuning of Vision-Language-Action Models through Chunked Offline Reinforcement Learning
  2. 2026Reinforcement Learning for Real-Time Vision-Language-Action Policies
  3. 2025Reinforcing Action Policies by Prophesying
  4. 2025RLinf-VLA: A Unified and Efficient Framework for Reinforcement Learning of Vision-Language-Action Models
  5. 2026IG-RFT: An Interaction-Guided RL Framework for VLA Models in Long-Horizon Robotic Manipulation
  6. 2025Dual-Actor Fine-Tuning of VLA Models: A Talk-and-Tweak Human-in-the-Loop Approach
  7. 2025World-Env: Leveraging World Model as a Virtual Environment for VLA Post-Training
  8. 2025VLA-R1: Enhancing Reasoning in Vision-Language-Action Models