Embodied-R: Collaborative Framework for Activating Embodied Spatial Reasoning in Foundation Models via Reinforcement Learning
Authors:
Preprint
Abstract
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.Keywords
All Related Versions
This publication has 22 references indexed in Scilit:
- A Comprehensive Survey of Small Language Models in the Era of Large Language Models: Techniques, Enhancements, Applications, Collaboration with LLMs, and TrustworthinessACM Transactions on Intelligent Systems and Technology, 2025
- Aligning Cyber Space With Physical World: A Comprehensive Survey on Embodied AIIEEE/ASME Transactions on Mechatronics, 2025
- Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces2025
- Visual Spatial ReasoningTransactions of the Association for Computational Linguistics, 2023
- MathPrompter: Mathematical Reasoning using Large Language Models2023
- Towards artificial general intelligence via a multimodal foundation modelNature Communications, 2022
- ScanQA: 3D Question Answering for Spatial Scene Understanding2022
- Foundations of human reasoning in the prefrontal cortexScience, 2014
- The Spatial Resolution of Visual AttentionCognitive Psychology, 2001
- Occipital cortex in man: Organization of callosal connections, related myelo‐ and cytoarchitecture, and putative boundaries of functional visual areasJournal of Comparative Neurology, 1990