LLMs as MDP Designers: A Dependency-Aware Agentic Framework for Automated Robot Policy Generation
LLMs as MDP Designers: A Dependency-Aware Agentic Framework for Automated Robot Policy Generation
This paper introduces ARPG, a dependency-aware agentic framework that leverages large language models (LLMs) not as execution-time decision makers but as designers to construct reinforcement learning pipelines before deployment. The authors investigate automated generation of executable robot RL policies from natural-language task descriptions by addressing the phase-wise structure of MDP formulation and simulator implementation. Experiments in Isaac Sim show that ARPG improves modeling correctness, environment executability, and policy-generation success compared with monolithic LLM baselines and off-the-shelf agentic systems.
The study presents ARPG, a framework that uses LLMs to design robot RL pipelines by structuring MDP formulation and simulator implementation sequentially, demonstrating improved modeling correctness and policy-generation success in Isaac Sim compared to baselines.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
artifacts. ARPG targets a key failure mode of automated policy generation: executable code may still yield non-learnable policies when state, action, reward, and termination artifacts are not semantically aligned. Experiments in Isaac Sim show that ARPG improves modeling correctness, environment executability, and policy-generation success compared with monolithic LLM baselines and off-the-shelf agentic systems. Add
Open source S4
arXiv:2611.04974v1 Announce Type: new Abstract: We investigate the automated generation of executable robot reinforcement learning (RL) policies from natural-language task descriptions. Rather than using large language models (LLMs) as execution-time decision makers, we use them as designers that construct RL pipelines before deployment, including Markov decision process (MDP) formulation and simulator implementation
Open source S2
Source：arXiv Robotics — research abstracts · arxiv.org