mlsanityData to 14 Oct 2026

Embodied-R: Collaborative Framework for Activating Embodied Spatial Reasoning in Foundation Models via Reinforcement Learning

arXiv 2504.12680 ↗First posted 17 Apr 2025cs.AI cs.CVcited by 37

On the map: LLM reasoning → Spatial and embodied reasoning

Abstract (arXiv)

Humans can perceive and reason about spatial relationships from sequential visual observations, such as egocentric video streams. However, how pretrained models acquire such abilities, especially high-level reasoning, remains unclear. This paper introduces Embodied-R, a collaborative framework combining large-scale Vision-Language Models (VLMs) for perception and small-scale Language Models (LMs) for reasoning. Using Reinforcement Learning (RL) with a novel reward system considering think-answer logical consistency, the model achieves slow-thinking capabilities with limited computational resources. After training on only 5k embodied video samples, Embodied-R with a 3B LM matches state-of-the-art multimodal reasoning models (OpenAI-o1, Gemini-2.5-pro) on both in-distribution and out-of-distribution embodied spatial reasoning tasks. Embodied-R also exhibits emergent thinking patterns such as systematic analysis and contextual integration. We further explore research questions including response length, training on VLM, strategies for reward design, and differences in model generalization after SFT (Supervised Fine-Tuning) and RL training.

Key facts. Cited by 37 papers in the map (data through 14 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 27 of its 64 references that are in the map. First posted on arXiv 17 Apr 2025.

Builds on

27 of 64 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces18 Dec
  2. 2024VidEgoThink: Assessing Egocentric Video Understanding Capabilities for Embodied AI15 Oct

Other references · 25

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  3. 2025Qwen2.5-VL Technical Reportcited by 5,587
  4. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,917
  5. 2023Video-LLaVA: Learning United Visual Representation by Alignment Before Projectioncited by 1,402
  6. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  7. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,090
  8. 2023Gemini: A Family of Highly Capable Multimodal Modelscited by 872
  9. 2025Visual-RFT: Visual Reinforcement Fine-Tuningcited by 513
  10. 2025rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinkingcited by 313
  11. 2025Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learningcited by 262
  12. 2023MathPrompter: Mathematical Reasoning using Large Language Modelscited by 236
13 more
  1. 2024Aligning Cyber Space with Physical World: A Comprehensive Survey on Embodied AIcited by 179
  2. 2024Imitate, Explore, and Self-Improve: A Reproduction Report on Slow-thinking Reasoning Systemscited by 164
  3. 2024Video-of-Thought: Step-by-Step Video Reasoning from Perception to Cognitioncited by 153
  4. 2025LlamaV-o1: Rethinking Step-by-step Visual Reasoning in LLMscited by 147
  5. 2024A Comprehensive Survey of Small Language Models in the Era of Large Language Models: Techniques, Enhancements, Applications, Collaboration with LLMs, and Trustworthinesscited by 127
  6. 2024AutoRT: Embodied Foundation Models for Large Scale Orchestration of Robotic Agentscited by 87
  7. 2023Effective Prompt Extraction from Language Modelscited by 85
  8. 2024Calibrating Large Language Models Using Their Generations Onlycited by 65
  9. 2024EmbodiedCity: A Benchmark Platform for Embodied Agent in Real-world City Environmentcited by 42
  10. 2025UrbanVideo-Bench: Benchmarking Vision-Language Models on Embodied Intelligence with Video Data in Urban Spacescited by 36
  11. 2024Purifying Large Language Models by Ensembling a Small Language Modelcited by 25
  12. 2025video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Modelcited by 24
  13. 2024AlanaVLM: A Multimodal Embodied AI Foundation Model for Egocentric Video Understandingcited by 23

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 26; 2026: 11.

37 papers.

Sort:

By area:

  1. 2025DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learningcited by 410
  2. 2025The Landscape of Agentic Reinforcement Learning for LLMs: A Surveycited by 205
  3. 2025Grounded Reinforcement Learning for Visual Reasoningcited by 83
  4. 2025Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Surveycited by 69
  5. 2025Look-Back: Implicit Visual Re-focusing in MLLM Reasoningcited by 55
  6. 2026A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agentscited by 50
  7. 2025How Far are VLMs from Visual Spatial Intelligence? A Benchmark-Driven Perspectivecited by 43
  8. 2025Multimodal Spatial Reasoning in the Large Model Era: A Survey and Benchmarkscited by 33
  9. 2025ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language Modelscited by 27
  10. 2025Embodied AI: From LLMs to World Modelscited by 24
  11. 2025Spatial Reasoning in Multimodal Large Language Models: A Survey of Tasks, Benchmarks and Methodscited by 23
  12. 2025WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learningcited by 21
25 more
  1. 2025WebSynthesis: World-Model-Guided MCTS for Efficient WebUI-Trajectory Synthesiscited by 21
  2. 2025Balanced Token Pruning: Accelerating Vision Language Models Beyond Local Optimizationcited by 20
  3. 2025DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Modelscited by 18
  4. 2025SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agentscited by 15
  5. 2025Reinforcement Fine-Tuning Powers Reasoning Capability of Multimodal Large Language Modelscited by 14
  6. 2025SpinBench: Perspective and Rotation as a Lens on Spatial Reasoning in VLMscited by 14
  7. 2025Reinforced Reasoning for Embodied Planningcited by 12
  8. 2025Reinforcement Learning for Large Model: A Surveycited by 9
  9. 2025AirScape: An Aerial Generative World Model with Motion Controllabilitycited by 7
  10. 2025CoNav: Collaborative Cross-Modal Reasoning for Embodied Navigationcited by 5
  11. 2025EgoPrune: Efficient Token Pruning for Egomotion Video Reasoning in Embodied Agentcited by 4
  12. 2025Video2Layout: Recall and Reconstruct Metric-Grounded Cognitive Map for Spatial Reasoningcited by 4
  13. 2025ReEXplore: Improving MLLMs for Embodied Exploration with Contextualized Retrospective Experience Replaycited by 4
  14. 2026ESPIRE: A Diagnostic Benchmark for Embodied Spatial Reasoning of Vision-Language Modelscited by 3
  15. 2026Are Video Reasoning Models Ready to Go Outside?cited by 2
  16. 2026DRIVESPATIAL: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Drivingcited by 2
  17. 2025Towards Accurate UAV Image Perception: Guiding Vision-Language Models with Stronger Task Promptscited by 1
  18. 2025From Macro to Micro: Benchmarking Microscopic Spatial Intelligence on Molecules via Vision-Language Modelscited by 1
  19. 20263D-RFT: Reinforcement Fine-Tuning for Video-based 3D Scene Understandingcited by 1
  20. 2026Rethinking Agentic Reinforcement Learning In Large Language Modelscited by 1
  21. 2026How Far Are Large Multimodal Models from Human-Level Spatial Action? A Benchmark for Goal-Oriented Embodied Navigation in Urban Airspacecited by 0
  22. 2026UniETP: Unifying Environments for Generalizable Embodied Task Planningcited by 0
  23. 2026SpatialCORE: Confidence-Aware Grounded Spatial Reasoning in Large Vision--Language Modelscited by 0
  24. 2026Soft Spatial Reasoningcited by 0
  25. 2026From Reasoning Failures to Composable Video Spatial Intelligencecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025BLM1: A Boundless Large Model for Cross-Space, Cross-Task, and Cross-Embodiment Learning
  2. 2025EmbodiedVSR: Dynamic Scene Graph-Guided Chain-of-Thought Reasoning for Visual Spatial Tasks
  3. 2025REM: Evaluating LLM Embodied Spatial Reasoning through Multi-Frame Trajectories
  4. 2025Embodied-Reasoner: Synergizing Visual Search, Reasoning, and Action for Embodied Interactive Tasks
  5. 2025ST-Think: How Multimodal Large Language Models Reason About 4D Worlds from Ego-Centric Videos
  6. 2024EmbSpatial-Bench: Benchmarking Spatial Understanding for Embodied Tasks with Large Vision-Language Models
  7. 2025Robot-R1: Reinforcement Learning for Enhanced Embodied Reasoning in Robotics
  8. 2026HY-Embodied-0.5: Embodied Foundation Models for Real-World Agents