Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2412.09666 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912154021527552 |
|---|---|
| author | Li, Haoming Chen, Zhaoliang Liu, Songyuan Lu, Yiming Liu, Fei |
| author_facet | Li, Haoming Chen, Zhaoliang Liu, Songyuan Lu, Yiming Liu, Fei |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_09666 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Systematic Analysis of LLM Contributions to Planning: Solver, Verifier, Heuristic Li, Haoming Chen, Zhaoliang Liu, Songyuan Lu, Yiming Liu, Fei Artificial Intelligence Computation and Language 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. |
| title | Systematic Analysis of LLM Contributions to Planning: Solver, Verifier, Heuristic |
| topic | Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2412.09666 |