optimization

OptGraph: Large Language Models Enhanced Evolutionary Optimization Via Graph Retrieval-Augmented Generation

arXiv:2607.27918

summary

The paper introduces OptGraph, a system that enhances automated evolutionary optimization by using large language models together with a graph‑based retrieval‑augmented generation approach to store and reuse structured experience for better pattern reuse and error‑aware refinement.

Abstract

Large language models (LLMs) have emerged as a powerful tool for automated evolutionary optimization, but existing methods remain limited in pattern reuse, error-aware refinement, and retrieval robustness across diverse tasks. To address these limitations, we propose OptGraph, the first optimization agentic workflow that introduces graph retrieval-augmented generation (GraphRAG). Specifically, OptGraph first constructs reusable experience as a typed graph, capturing the relationships among modeling patterns, problem formalization, implementation details, and error corrections. In the inference stage, OptGraph leverages graph neighborhood information to enrich retrieved knowledge, providing structured context to improve modeling, verification, and iterative refinement. Moreover, OptGraph supports adaptive knowledge updates, enabling the distillation of execution traces and verification feedback into reusable graph knowledge without ndertaking LLM parameter tuning. Extensive experiments on benchmark datasets show that our proposed OptGraph achieves an average exact accuracy 8.9% higher than the state-of-the-art prompt-based automated optimization frameworks. Our code has been made available at https://github.com/xianchaoxiu/OptGraph.

Topics & keywords

#evolutionary optimization#large language models#graph retrieval#knowledge graphs#automated algorithm designgraph retrieval-augmented generationtyped experience graphprompt-based optimizationLLM-driven evolutionary algorithmserror-aware refinement