Systematic Analysis of LLM Contributions to Planning: Solver, Verifier, Heuristic
This work addresses the challenge of integrating LLMs into planning systems for AI and robotics, offering incremental insights into optimizing their use as heuristic guides rather than direct solvers.
The paper systematically analyzes how large language models (LLMs) contribute to planning problems, finding that while LLMs struggle to generate correct plans directly, they excel at providing heuristic feedback to improve intermediate solutions, with potential applications in diverse reasoning tasks.
In this work, we provide a systematic analysis of how large language models (LLMs) contribute to solving planning problems. In particular, we examine how LLMs perform when they are used as problem solver, solution verifier, and heuristic guidance to improve intermediate solutions. Our analysis reveals that although it is difficult for LLMs to generate correct plans out-of-the-box, LLMs are much better at providing feedback signals to intermediate/incomplete solutions in the form of comparative heuristic functions. This evaluation framework provides insights into how future work may design better LLM-based tree-search algorithms to solve diverse planning and reasoning problems. We also propose a novel benchmark to evaluate LLM's ability to learn user preferences on the fly, which has wide applications in practical settings.