The Lighthouse of Language: Enhancing LLM Agents via Critique-Guided Improvement

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Yang, Ruihan, Ye, Fanghua, Li, Jian, Yuan, Siyu, Zhang, Yikai, Tu, Zhaopeng, Li, Xiaolong, Yang, Deqing
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914112798195712
author Yang, Ruihan
Ye, Fanghua
Li, Jian
Yuan, Siyu
Zhang, Yikai
Tu, Zhaopeng
Li, Xiaolong
Yang, Deqing
author_facet Yang, Ruihan
Ye, Fanghua
Li, Jian
Yuan, Siyu
Zhang, Yikai
Tu, Zhaopeng
Li, Xiaolong
Yang, Deqing
contents Large language models (LLMs) have recently transformed from text-based assistants to autonomous agents capable of planning, reasoning, and iteratively improving their actions. While numerical reward signals and verifiers can effectively rank candidate actions, they often provide limited contextual guidance. In contrast, natural language feedback better aligns with the generative capabilities of LLMs, providing richer and more actionable suggestions. However, parsing and implementing this feedback effectively can be challenging for LLM-based agents. In this work, we introduce Critique-Guided Improvement (CGI), a novel two-player framework, comprising an actor model that explores an environment and a critic model that generates detailed nature language feedback. By training the critic to produce fine-grained assessments and actionable revisions, and the actor to utilize these critiques, our approach promotes more robust exploration of alternative strategies while avoiding local optima. Experiments in three interactive environments show that CGI outperforms existing baselines by a substantial margin. Notably, even a small critic model surpasses GPT-4 in feedback quality. The resulting actor achieves state-of-the-art performance, demonstrating the power of explicit iterative guidance to enhance decision-making in LLM-based agents.
format Preprint
id arxiv_https___arxiv_org_abs_2503_16024
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Lighthouse of Language: Enhancing LLM Agents via Critique-Guided Improvement
Yang, Ruihan
Ye, Fanghua
Li, Jian
Yuan, Siyu
Zhang, Yikai
Tu, Zhaopeng
Li, Xiaolong
Yang, Deqing
Computation and Language
Artificial Intelligence
Large language models (LLMs) have recently transformed from text-based assistants to autonomous agents capable of planning, reasoning, and iteratively improving their actions. While numerical reward signals and verifiers can effectively rank candidate actions, they often provide limited contextual guidance. In contrast, natural language feedback better aligns with the generative capabilities of LLMs, providing richer and more actionable suggestions. However, parsing and implementing this feedback effectively can be challenging for LLM-based agents. In this work, we introduce Critique-Guided Improvement (CGI), a novel two-player framework, comprising an actor model that explores an environment and a critic model that generates detailed nature language feedback. By training the critic to produce fine-grained assessments and actionable revisions, and the actor to utilize these critiques, our approach promotes more robust exploration of alternative strategies while avoiding local optima. Experiments in three interactive environments show that CGI outperforms existing baselines by a substantial margin. Notably, even a small critic model surpasses GPT-4 in feedback quality. The resulting actor achieves state-of-the-art performance, demonstrating the power of explicit iterative guidance to enhance decision-making in LLM-based agents.
title The Lighthouse of Language: Enhancing LLM Agents via Critique-Guided Improvement
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2503.16024