Chapter 2 · 2025
Pre-Act: Multi-Step Planning and Reasoning Improves Acting in LLM Agents
Mrinal Rawat, Ambuje Gupta, Rushil Goomer
Abstract
The ReAct capability in LLMs forms the foundation of modern agentic systems, but smaller models struggle with complex reasoning tasks. This paper introduces Pre-Act, a novel approach that enhances agent performance by creating a multi-step execution plan with detailed reasoning. It incrementally incorporates previous steps and tool outputs, refining itself after each step. Experiments show that fine-tuned smaller models using Pre-Act can outperform larger models like GPT-4 in action accuracy and goal completion rate.
Topics
LLM agentsmulti-step planningreasoningReActtask-oriented agents
Relevance Scores
Long-Horizon Score85
Enterprise Score80
Completeness75
Paper Info
Year2025
Venue
Type
ChapterCh. 2
Authors3
Zone III Analysis
Frameworks
Related Papers
ReAct: Synergizing Reasoning and Acting in Language Mod…
2023 · Ch.1
Reflexion: Language Agents with Verbal Reinforcement Le…
2023 · Ch.1
Tree of Thoughts: Deliberate Problem Solving with Large…
2023 · Ch.1
Toolformer: Language Models Can Teach Themselves to Use…
2023 · Ch.1
View all Chapter 2 papers →