Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement

TL;DR

This paper introduces extsc{Multi-Agent Procedural Graph Extraction}, a multi-agent framework that iteratively refines procedural graphs from natural language by addressing structural and logical errors, significantly improving correctness and consistency.

Contribution

It proposes a novel multi-agent, multi-round reasoning framework with dedicated structural and logical refinement stages for extracting accurate procedural graphs from text.

Findings

Achieves substantial improvements in structural correctness.

Enhances logical consistency of extracted graphs.

Operates without supervision or parameter updates.

Abstract

Automatically extracting workflows as procedural graphs from natural language is promising yet underexplored, demanding both structural validity and logical alignment. While recent large language models (LLMs) show potential for procedural graph extraction, they often produce ill-formed structures or misinterpret logical flows. We present \model{}, a multi-agent framework that formulates procedural graph extraction as a multi-round reasoning process with dedicated structural and logical refinement. The framework iterates through three stages: (1) a graph extraction phase with the graph builder agent, (2) a structural feedback phase in which a simulation agent diagnoses and explains structural defects, and (3) a logical feedback phase in which a semantic agent aligns semantics between flow logic and linguistic cues in the source text. Important feedback is prioritized and expressed in natural language, which is injected into subsequent prompts, enabling interpretable and controllable refinement. This modular design allows agents to target distinct error types without supervision or parameter updates. Experiments demonstrate that \model{} achieves substantial improvements in both structural correctness and logical consistency over strong baselines.