# Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement

arXiv: https://arxiv.org/abs/2601.19170 · HTML: https://mlsanity.com/paper/2601.19170/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 24 of its 50 references that are in the map.
- First posted on arXiv 27 Jan 2026.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 24 of 24)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Unifying Large Language Models and Knowledge Graphs: A Roadmap](https://mlsanity.com/paper/2306.08302/) (2023)
- [A Survey on Knowledge Graphs: Representation, Acquisition and Applications](https://mlsanity.com/paper/2002.00388/) (2020)
- [End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF](https://mlsanity.com/paper/1603.01354/) (2016)
- [A Dataset for Tracking Entities in Open Domain Procedural Text](https://mlsanity.com/paper/2011.08092/) (2020)
- [Self-optimizing Feature Generation via Categorical Hashing Representation and Hierarchical Reinforcement Crossing](https://mlsanity.com/paper/2309.04612/) (2023)
- [Evolutionary Large Language Model for Automated Feature Transformation](https://mlsanity.com/paper/2405.16203/) (2024)
- [Feature Selection as Deep Sequential Generative Learning](https://mlsanity.com/paper/2403.03838/) (2024)
- [Unsupervised Generative Feature Transformation via Graph Contrastive Pre-training and Multi-objective Fine-tuning](https://mlsanity.com/paper/2405.16879/) (2024)
- [CT-PatchTST: Channel-Time Patch Time-Series Transformer for Long-Term Renewable Energy Forecasting](https://mlsanity.com/paper/2501.08620/) (2025)
- [PET: An Annotated Dataset for Process Extraction from Natural Language Text](https://mlsanity.com/paper/2203.04860/) (2022)
- [Title2Event: Benchmarking Open Event Extraction with a Large-scale Chinese Title Dataset](https://mlsanity.com/paper/2211.00869/) (2022)
- [Neuro-Symbolic Embedding for Short and Effective Feature Selection via Autoregressive Generation](https://mlsanity.com/paper/2404.17157/) (2024)
- [Revolutionizing Biomarker Discovery: Leveraging Generative AI for Bio-Knowledge-Embedded Continuous Space Exploration](https://mlsanity.com/paper/2409.15612/) (2024)
- [Topology-aware Reinforcement Feature Space Reconstruction for Graph Data](https://mlsanity.com/paper/2411.05742/) (2024)
- [Constructing Flow Graphs from Procedural Cybersecurity Texts](https://mlsanity.com/paper/2105.14357/) (2021)
- [Set-Aligning Framework for Auto-Regressive Event Temporal Graph Generation](https://mlsanity.com/paper/2404.01532/) (2024)
- [Efficient Post-Training Refinement of Latent Reasoning in Large Language Models](https://mlsanity.com/paper/2506.08552/) (2025)
- [PAGED: A Benchmark for Procedural Graphs Extraction from Documents](https://mlsanity.com/paper/2408.03630/) (2024)
- [Leveraging pre-trained language models for conversational information seeking from text](https://mlsanity.com/paper/2204.03542/) (2022)
- [Cascading Large Language Models for Salient Event Graph Generation](https://mlsanity.com/paper/2406.18449/) (2024)
- [Data-Efficient Symbolic Regression via Foundation Model Distillation](https://mlsanity.com/paper/2508.19487/) (2025)

## Abstract (arXiv)

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.