mlsanityData to 14 Oct 2026

Embodied-R1: Reinforced Embodied Reasoning for General Robotic Manipulation

arXiv 2508.13998 ↗First posted 19 Aug 2025cs.RO cs.AI cs.LGcited by 38

On the map: LLM reasoning → Spatial and embodied reasoning

Abstract (arXiv)

Generalization in embodied AI is hindered by the "seeing-to-doing gap," which stems from data scarcity and embodiment heterogeneity. To address this, we pioneer "pointing" as a unified, embodiment-agnostic intermediate representation, defining four core embodied pointing abilities that bridge high-level vision-language comprehension with low-level action primitives. We introduce Embodied-R1, a 3B Vision-Language Model (VLM) specifically designed for embodied reasoning and pointing. We use a wide range of embodied and general visual reasoning datasets as sources to construct a large-scale dataset, Embodied-Points-200K, which supports key embodied pointing capabilities. We then train Embodied-R1 using a two-stage Reinforced Fine-tuning (RFT) curriculum with a specialized multi-task reward design. Embodied-R1 achieves state-of-the-art performance on 11 embodied spatial and pointing benchmarks. Critically, it demonstrates robust zero-shot generalization by achieving a 56.2% success rate in the SIMPLEREnv and 87.5% across 8 real-world XArm tasks without any task-specific fine-tuning, representing a 62% improvement over strong baselines. Furthermore, the model exhibits high robustness against diverse visual disturbances. Our work shows that a pointing-centric representation, combined with an RFT training paradigm, offers an effective and generalizable pathway to closing the perception-action gap in robotics.

Key facts. Cited by 38 papers in the map (data through 14 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 61 of its 84 references that are in the map. First posted on arXiv 19 Aug 2025.

Builds on

61 of 84 references.

Closest ancestors · 12

Same direction, newest first

  1. 2025Affordance-R1: Reinforcement Learning for Generalizable Affordance Reasoning in Multimodal Large Language Model8 Aug
  2. 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning22 Jul
  3. 2025RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics4 Jun
  4. 2025From Seeing to Doing: Bridging Reasoning and Decision for Robotic Manipulation13 May
  5. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success27 Feb
  6. 2025SoFar: Language-Grounded Orientation Bridges Spatial Reasoning and Object Manipulation18 Feb
  7. 2025SpatialCoT: Advancing Spatial Reasoning through Coordinate Alignment and Chain-of-Thought for Embodied Task Planning17 Jan
  8. 2024ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation3 Sep
  9. 2024Robotic Control via Embodied Chain-of-Thought Reasoning11 Jul
  10. 2024TopViewRS: Vision-Language Models as Top-View Spatial Reasoners4 Jun
  11. 2024SpatialRGPT: Grounded Spatial Reasoning in Vision Language Models3 Jun
  12. 2023What's "up" with vision-language models? Investigating their struggle with spatial reasoning30 Oct

Other references · 49

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  4. 2025Qwen2.5-VL Technical Reportcited by 5,587
  5. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,373
  6. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,887
  7. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  8. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,407
  9. 2024OpenAI o1 System Cardcited by 1,244
  10. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  11. 2024Grounded SAM: Assembling Open-World Models for Diverse Visual Taskscited by 929
  12. 2023Open X-Embodiment: Robotic Learning Datasets and RT-X Modelscited by 862
37 more
  1. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 775
  2. 2023VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Modelscited by 750
  3. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  4. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 665
  5. 2024Evaluating Real-World Robot Manipulation Policies in Simulationcited by 495
  6. 2025Video-R1: Reinforcing Video Reasoning in MLLMscited by 484
  7. 2024Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Modelscited by 329
  8. 2025VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learningcited by 323
  9. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 304
  10. 2024RoboPoint: A Vision-Language Model for Spatial Affordance Prediction for Roboticscited by 251
  11. 2025Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive Reinforcementcited by 239
  12. 2023Any-point Trajectory Modeling for Policy Learningcited by 215
  13. 2025VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learningcited by 176
  14. 2025ConRFT: A Reinforced Fine-tuning Method for VLA Models via Consistency Policycited by 154
  15. 2024Data Scaling Laws in Imitation Learning for Robotic Manipulationcited by 145
  16. 2025Magma: A Foundation Model for Multimodal AI Agentscited by 139
  17. 2024RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Roboticscited by 136
  18. 2025RoboBrain: A Unified Brain Model for Robotic Manipulation from Abstract to Concretecited by 135
  19. 2023RT-Trajectory: Robotic Task Generalization via Hindsight Trajectory Sketchescited by 133
  20. 2024Flow as the Cross-Domain Manipulation Interfacecited by 132
  21. 2025HAMSTER: Hierarchical Action Models For Open-World Robot Manipulationcited by 130
  22. 2024MOKA: Open-World Robotic Manipulation through Mark-Based Visual Promptingcited by 129
  23. 2024CoTracker3: Simpler and Better Point Tracking by Pseudo-Labelling Real Videoscited by 128
  24. 2024Track2Act: Predicting Point Tracks from Internet Videos enables Generalizable Robot Manipulationcited by 124
  25. 2024SAT: Dynamic Spatial Aptitude Training for Multimodal Language Modelscited by 114
  26. 2025What Can RL Bring to VLA Generalization? An Empirical Studycited by 108
  27. 2024EmbSpatial-Bench: Benchmarking Spatial Understanding for Embodied Tasks with Large Vision-Language Modelscited by 107
  28. 2025RoboBrain 2.0 Technical Reportcited by 89
  29. 2024General Flow as Foundation Affordance for Scalable Robot Learningcited by 85
  30. 2025DeepVideo-R1: Video Reinforcement Fine-Tuning via Difficulty-aware Regressive GRPOcited by 61
  31. 2025VisionReasoner: Unified Reasoning-Integrated Visual Perception via Reinforcement Learningcited by 51
  32. 2025A0: An Affordance-Aware Hierarchical Model for General Robotic Manipulationcited by 45
  33. 2023AlignDiff: Aligning Diverse Human Preferences via Behavior-Customisable Diffusion Modelcited by 37
  34. 2024A3VLM: Actionable Articulation-Aware Vision Language Modelcited by 33
  35. 2024Uni-RLHF: Universal Platform and Benchmark Suite for Reinforcement Learning with Diverse Human Feedbackcited by 21
  36. 2024Enhancing Robotic Manipulation with AI Feedback from Multimodal Large Language Modelscited by 10
  37. 2024CoReEcho: Continuous Representation Learning for 2D+time Echocardiography Analysiscited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 11; 2026: 27.

38 papers.

Sort:

By area:

  1. 2025FlowVLA: Visual Chain of Thought-based Motion Reasoning for Vision-Language-Action Modelscited by 50
  2. 2025Pure Vision Language Action (VLA) Models: A Comprehensive Surveycited by 49
  3. 2026Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planningcited by 24
  4. 2025A Survey on Agentic Multimodal Large Language Modelscited by 23
  5. 2025Vlaser: Vision-Language-Action Model with Synergistic Embodied Reasoningcited by 23
  6. 2025TraceGen: World Modeling in 3D Trace Space Enables Learning from Cross-Embodiment Videoscited by 15
  7. 2026Not All Features Are Created Equal: A Mechanistic Study of Vision-Language-Action Modelscited by 15
  8. 2025Towards Spatial Trace with Reasoning in Vision-Language Models for Roboticscited by 12
  9. 2025ERA: Transforming VLMs into Embodied Agents via Embodied Prior Learning and Online Reinforcement Learningcited by 10
  10. 2025Pelican-VL 1.0: A Foundation Brain Model for Embodied Intelligencecited by 10
  11. 2025PhysToolBench: Benchmarking Physical Tool Understanding for MLLMscited by 9
  12. 2026Long-Horizon Manipulation via Trace-Conditioned VLA Planningcited by 9
26 more
  1. 2026ProcVLM: Learning Procedure-Grounded Progress Rewards for Robotic Manipulationcited by 9
  2. 2025LocateAnything3D: Vision-Language 3D Detection with Chain-of-Sightcited by 8
  3. 2026Revisiting Embodied Chain-of-Thought for Generalizable Robot Manipulationcited by 7
  4. 2026VLA-Thinker: Boosting Vision-Language-Action Models through Thinking-with-Image Reasoningcited by 6
  5. 2026GEM: Generative Supervision Helps Embodied Intelligencecited by 6
  6. 2026Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Modelscited by 6
  7. 2026ForceFlow: Learning to Feel and Act via Contact-Driven Flow Matchingcited by 5
  8. 2025Universal Dexterous Functional Grasping via Demonstration-Editing Reinforcement Learningcited by 4
  9. 2026VISTA: Enhancing Visual Conditioning via Track-Following Preference Optimization in Vision-Language-Action Modelscited by 3
  10. 2026ST-VLA: Enabling 4D-Aware Spatiotemporal Understanding for General Robot Manipulationcited by 3
  11. 2026Interact3D: Compositional 3D Generation of Interactive Objectscited by 3
  12. 2026RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Modelscited by 3
  13. 2026Dense Reward for Multi-View 3D Reasoning with Global Maps and Local Viewscited by 3
  14. 2026XEmbodied: A Foundation Model with Enhanced Geometric and Physical Cues for Large-Scale Embodied Environmentscited by 2
  15. 2026VISTA: Vision-Grounded and Physics-Validated Adaptation of UMI data for VLA Trainingcited by 2
  16. 2026ISO: An RLVR-Native Optimization Stackcited by 2
  17. 2026Pointing-VLA: Typed Spatial Grounding Interfaces for Vision-Language-Action Manipulationcited by 2
  18. 20263D-RFT: Reinforcement Fine-Tuning for Video-based 3D Scene Understandingcited by 1
  19. 2026Spatially Grounded Long-Horizon Task Planning in the Wildcited by 1
  20. 2026E-TTS: A New Embodied Test-Time Scaling Framework for Robotic Manipulationcited by 1
  21. 2026Mitigating State Aliasing in Vision-Language-Action Models via Inverse Dynamics Learningcited by 0
  22. 2026RoboPIN: Grounded Embodied Reasoning via Pinned Chain-of-Thoughtcited by 0
  23. 2026Exo2EgoPose: Leveraging Exocentric Demonstrations for Vision-Language guided Egocentric 3D Hand Pose Forecastingcited by 0
  24. 2026Capek 0.5: An Execution-Centric Vision-Language Model for Embodied Intelligencecited by 0
  25. 2026PhysBrain 1.5: From Vision-Language Models to Physical Foundation Modelscited by 0
  26. 2026EgoLAP: Learning from Egocentric Human Data through Language-Action Reasoningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Robot-R1: Reinforcement Learning for Enhanced Embodied Reasoning in Robotics
  2. 2026Technical Report of RoboSpatial Challenge at CVPR 2026: Selective Reasoning Activation and Reference-Frame Disambiguation for Embodied Spatial Reasoning
  3. 2025Unified Embodied VLM Reasoning with Robotic Action via Autoregressive Discretized Pre-training
  4. 2025Embodied4C: Measuring What Matters for Embodied Vision-Language Navigation
  5. 2025Mind to Hand: Purposeful Robotic Control via Embodied Reasoning
  6. 2025BLM1: A Boundless Large Model for Cross-Space, Cross-Task, and Cross-Embodiment Learning
  7. 2024Emma-X: An Embodied Multimodal Action Model with Grounded Chain of Thought and Look-ahead Spatial Reasoning
  8. 2026Guava: Distilling Frontier VLMs into a Compact Agent through a Robotic Manipulation Harness