mlsanityData to 7 Oct 2026

MolmoAct: Action Reasoning Models that can Reason in Space

arXiv 2508.07917 ↗First posted 11 Aug 2025cs.ROcited by 189

On the map: LLM reasoning → Spatial and embodied reasoning

Abstract (arXiv)

Reasoning is central to purposeful action, yet most robotic foundation models map perception and instructions directly to control, which limits adaptability, generalization, and semantic grounding. We introduce Action Reasoning Models (ARMs), a class of robotic foundation models that integrate perception, planning, and control through a structured three-stage pipeline. Our model, MolmoAct, encodes observations and instructions into depth-aware perception tokens, generates mid-level spatial plans as editable trajectory traces, and predicts precise low-level actions, enabling explainable and steerable behavior. MolmoAct-7B-D achieves strong performance across simulation and real-world settings: 70.5% zero-shot accuracy on SimplerEnv Visual Matching tasks, surpassing closed-source Pi-0 and GR00T N1.5; 86.6% average success on LIBERO, including an additional 6.3% gain over ThinkAct on long-horizon tasks; and in real-world fine-tuning, an additional 10% (single-arm) and an additional 22.7% (bimanual) task progression over Pi-0-FAST. It also outperforms baselines by an additional 23.3% on out-of-distribution generalization and achieves top human-preference scores for open-ended instruction following and trajectory steering. Furthermore, we release, for the first time, the MolmoAct Dataset -- a mid-training robot dataset comprising over 10,000 high quality robot trajectories across diverse scenarios and tasks. Training with this dataset yields an average 5.5% improvement in general performance over the base model. We release all model weights, training code, our collected dataset, and our action reasoning dataset, establishing MolmoAct as both a state-of-the-art robotics foundation model and an open blueprint for building ARMs that transform perception into purposeful action through structured reasoning. Blogpost: https://allenai.org/blog/molmoact

Key facts. Cited by 189 papers in the map (data through 7 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 31 of its 97 references that are in the map. First posted on arXiv 11 Aug 2025.

Builds on

31 of 97 references.

Closest ancestors · 8

Same direction, newest first

  1. 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning22 Jul
  2. 2025NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks28 Apr
  3. 2025CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models27 Mar
  4. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success27 Feb
  5. 2024Emma-X: An Embodied Multimodal Action Model with Grounded Chain of Thought and Look-ahead Spatial Reasoning16 Dec
  6. 2024AHA: A Vision-Language-Model for Detecting and Reasoning Over Failures in Robotic Manipulation1 Oct
  7. 2024ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation3 Sep
  8. 2024Robotic Control via Embodied Chain-of-Thought Reasoning11 Jul

Other references · 23

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  4. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  5. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  6. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  7. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,881
  8. 2022ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoningcited by 1,449
  9. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  10. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 895
  11. 2023Multimodal Chain-of-Thought Reasoning in Language Modelscited by 740
  12. 2024RDT-1B: a Diffusion Foundation Model for Bimanual Manipulationcited by 670
11 more
  1. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 643
  2. 2022BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learningcited by 517
  3. 2024CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulationcited by 411
  4. 2025WorldVLA: Towards Autoregressive Action World Modelcited by 351
  5. 2022Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoningcited by 347
  6. 2024TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulationcited by 331
  7. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 298
  8. 2025DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Controlcited by 191
  9. 2025Multimodal Chain-of-Thought Reasoning: A Comprehensive Surveycited by 176
  10. 2025HAMSTER: Hierarchical Action Models For Open-World Robot Manipulationcited by 125
  11. 2024Perception Tokens Enhance Visual Reasoning in Multimodal Language Modelscited by 68

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 43; 2026: 146.

189 papers.

Sort:

By area:

  1. 2026World Action Models are Zero-shot Policiescited by 411
  2. 2025LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorizationcited by 166
  3. 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Modelcited by 140
  4. 2026Cosmos 3: Omnimodal World Models for Physical AIcited by 137
  5. 2026Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Groundingcited by 132
  6. 2026ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learningcited by 94
  7. 2025InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policycited by 83
  8. 2025RynnVLA-002: A Unified Vision-Language-Action and World Modelcited by 70
  9. 2026Being-H0.7: A Latent World-Action Model from Egocentric Videoscited by 68
  10. 2026Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalizationcited by 61
  11. 2026Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisonscited by 51
  12. 2025VLA-0: Building State-of-the-Art VLAs with Zero Modificationcited by 49
38 more
  1. 2026MolmoAct2: Action Reasoning Models for Real-world Deploymentcited by 45
  2. 2025DepthVLA: Enhancing Vision-Language-Action Models with Depth-Aware Spatial Reasoningcited by 44
  3. 2026Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Executioncited by 42
  4. 2025NovaFlow: Zero-Shot Manipulation via Actionable Flow from Generated Videoscited by 38
  5. 2025Unified Diffusion VLA: Vision-Language-Action Model via Joint Discrete Denoising Diffusion Processcited by 37
  6. 2026Harness VLA: Steering Frozen VLAs into Reliable Manipulation Primitives via Memory-Guided Agentscited by 37
  7. 2025The Better You Learn, The Smarter You Prune: Towards Efficient Vision-language-action Models via Differentiable Token Pruningcited by 34
  8. 2025Don't Blind Your VLA: Aligning Visual Representations for OOD Generalizationcited by 34
  9. 2025FailSafe: Reasoning and Recovery from Failures in Vision-Language-Action Modelscited by 33
  10. 2026ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?cited by 33
  11. 2025EO-1: An Open Unified Embodied Foundation Model for General Robot Controlcited by 30
  12. 2026TOPReward: Token Probabilities as Hidden Zero-Shot Rewards for Roboticscited by 29
  13. 2026Interactive World Simulator for Robot Policy Training and Evaluationcited by 27
  14. 2025NORA-1.5: A Vision-Language-Action Model Trained using World Model- and Action-based Preference Rewardscited by 26
  15. 2025SRPO: Self-Referential Policy Optimization for Vision-Language-Action Modelscited by 25
  16. 2025Robobench: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models as Embodied Braincited by 24
  17. 2026Xiaomi-Robotics-1: Scaling Vision-Language-Action Models with over 100K Hours of Real-World Trajectoriescited by 24
  18. 2025Vlaser: Vision-Language-Action Model with Synergistic Embodied Reasoningcited by 22
  19. 2026AtomVLA: Scalable Post-Training for Robotic Manipulation via Predictive Latent World Modelscited by 21
  20. 2026Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planningcited by 20
  21. 2026A Systematic Study of Data Modalities and Strategies for Co-training Large Behavior Models for Robot Manipulationcited by 20
  22. 2025A Survey on Agentic Multimodal Large Language Modelscited by 19
  23. 2026Latent Reasoning VLA: Latent Thinking and Prediction for Vision-Language-Action Modelscited by 19
  24. 2026Steerable Vision-Language-Action Policies for Embodied Reasoning and Hierarchical Controlcited by 19
  25. 2026HY-Embodied-0.5: Embodied Foundation Models for Real-World Agentscited by 18
  26. 2026RHO: Your Coding Agent is Secretly a Roboticistcited by 18
  27. 2026Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoningcited by 17
  28. 2026LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfercited by 17
  29. 2026ActionCodec: What Makes for Good Action Tokenizerscited by 17
  30. 2025AsyncVLA: Asynchronous Flow Matching for Vision-Language-Action Modelscited by 16
  31. 2026Vision-aligned Latent Reasoning for Multi-modal Large Language Modelcited by 16
  32. 2025XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representationscited by 15
  33. 2025Mull-Tokens: Modality-Agnostic Latent Thinkingcited by 15
  34. 2026MaskWAM: Unifying Mask Prompting and Prediction for World-Action Modelscited by 15
  35. 2026TwinBrainVLA: Unleashing the Potential of Generalist VLMs for Embodied Tasks via Asymmetric Mixture-of-Transformerscited by 14
  36. 2025TraceGen: World Modeling in 3D Trace Space Enables Learning from Cross-Embodiment Videoscited by 13
  37. 2025Towards Spatial Trace with Reasoning in Vision-Language Models for Roboticscited by 12
  38. 2026Robotic Video World Models: A Survey of Applications, Research Challenges, Future Directionscited by 12

50 most cited of 189 All 189 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026ESPIRE: A Diagnostic Benchmark for Embodied Spatial Reasoning of Vision-Language Models
  2. 2025Seeing Across Views: Benchmarking Spatial Reasoning of Vision-Language Models in Robotic Scenes
  3. 2026RoBoSR: Structured Scene Representations for Embodied Robotic Reasoning ↗
  4. 2025SIMPACT: Simulation-Enabled Action Planning using Vision-Language Models
  5. 2025Do What You Say: Steering Vision-Language-Action Models via Runtime Reasoning-Action Alignment Verification
  6. 2026Capek 0.5: An Execution-Centric Vision-Language Model for Embodied Intelligence
  7. 2026HALO: A Unified Vision-Language-Action Model for Embodied Multimodal Chain-of-Thought Reasoning
  8. 2025Unified Embodied VLM Reasoning with Robotic Action via Autoregressive Discretized Pre-training