Textualized Agent-Style Reasoning for Complex Tasks by Multiple Round LLM Generation

TL;DR

AgentCOT is a novel autonomous agent framework that enhances complex reasoning in large language models through multiple rounds of generation, supporting interpretability and reducing hallucinations.

Contribution

We introduce AgentCOT, a new agent-style reasoning framework with graph-structured inference and strategies to improve LLM performance on complex tasks.

Findings

Significant performance improvements on six benchmarks.

Effective reduction of hallucinations and increased interpretability.

Enhanced reasoning capabilities through graph-structured logic.

Abstract

Chain-of-thought prompting significantly boosts the reasoning ability of large language models but still faces three issues: hallucination problem, restricted interpretability, and uncontrollable generation. To address these challenges, we present AgentCOT, a llm-based autonomous agent framework, which can solve complex problems in an agent-style manner by multiple round LLM generation. At each step, AgentCOT selects an action and executes it to yield an intermediate result with supporting evidence. In addition, we integrate the step's index into the reasoning process to form a graph structure for complex inference logic. We introduce two new strategies to enhance the performance of AgentCOT.We conduct extensive experiments to verify the effectiveness of our method on six common benchmarks. Results exhibit that our method brings in substantial improvements over current competitive approaches.