Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement

2026-01-27

AI Summary

Automatically extracting structurally valid and logically consistent flowcharts from natural language remains challenging, as current large language models often produce diagrams with structural errors or semantic misalignments. This work proposes a multi-agent collaborative framework that integrates a graph constructor, a structure simulator, and a semantic aligner to iteratively refine flowcharts through three stages—graph construction, structural feedback, and logical feedback—driven solely by natural language feedback, without requiring supervision or parameter updates. The approach enables interpretable and controllable flowchart refinement. Experimental results demonstrate that the method significantly outperforms strong baselines on metrics of structural correctness and logical consistency, confirming its effectiveness for automatic flowchart extraction from natural language.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Graph-based Machine Learning
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Agentic search

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.

Problem

• procedural graph extraction
• structural validity
• logical alignment
• multi-agent framework
• natural language processing

Innovation

• multi-agent framework
• procedural graph extraction
• structural refinement
• logical alignment
• interpretable refinement