Focused ReAct: Improving ReAct through Reiterate and Early Stop

TL;DR

Focused ReAct enhances the original ReAct method for large language models by incorporating reiteration and early stopping, significantly improving accuracy and reducing runtime in complex reasoning tasks.

Contribution

This paper introduces Focused ReAct, a novel approach that addresses focus loss and action loops in ReAct through reiteration and early stop mechanisms.

Findings

Accuracy improved by 18% to 530%.

Runtime reduced by up to 34%.

Effectiveness demonstrated on complex reasoning tasks.

Abstract

Large language models (LLMs) have significantly improved their reasoning and decision-making capabilities, as seen in methods like ReAct. However, despite its effectiveness in tackling complex tasks, ReAct faces two main challenges: losing focus on the original question and becoming stuck in action loops. To address these issues, we introduce Focused ReAct, an enhanced version of the ReAct paradigm that incorporates reiteration and early stop mechanisms. These improvements help the model stay focused on the original query and avoid repetitive behaviors. Experimental results show accuracy gains of 18% to 530% and a runtime reduction of up to 34% compared to the original ReAct method.