LLMs as MDP Designers: A Dependency-Aware Agentic Framework for Automated Robot Policy Generation
cs.RO, cs.AI
Submitted: 2026-11-03
Updated: 2026-11-03
Terminology
Sources
- Large Language Models for Robotics: A Survey
- Learning Agile Robotic Locomotion Skills by Imitating Animals
- Evaluating Large Language Models Trained on Code
- RT-1: Robotics Transformer for Real-World Control at Scale
- A Generalist Agent
- OpenVLA: An Open-Source Vision-Language-Action Model
- Taxonomy and Trends in Reinforcement Learning for Robotics and Control Systems: A Structured Review
- REBEL: Reward Regularization-Based Approach for Robotic Reinforcement Learning from Human Feedback
- Eureka: Human-Level Reward Design via Coding Large Language Models
- RoboReward: General-Purpose Vision-Language Reward Models for Robotics
- OptiMUS-0.3: Using Large Language Models to Model and Solve Optimization Problems at Scale
- Towards a Science of Scaling Agent Systems
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation