AutoSG: LLM-Driven Solver Generation Solely from Task Prompts for Expensive Optimization
2026-05-25 • Computation and Language
Computation and LanguageArtificial Intelligence
AI summaryⓘ
The authors developed AutoSG, a system that automatically creates specialized solvers for hard optimization problems by turning natural language instructions into working code. They addressed common issues with previous AI methods by using verified information from scientific literature, improving the solver in one careful step to keep important parts intact, and evaluating solvers without needing to test on real problem examples. Their tests show AutoSG works better than current human-made and AI-generated solvers on various challenging tasks. This approach helps make solver creation faster and more reliable.
OptimizationLarge Language Models (LLMs)Automated Solver GenerationSelf-RefinementRetrieval-Augmented GenerationElo Rating SystemExpensive OptimizationCode GenerationInstance-Free Evaluation
Authors
Haoran Gu, Handing Wang, Yi Mei, Mengjie Zhang
Abstract
Expensive optimization tasks are ubiquitous in real-world applications, demanding highly specialized solvers. While LLM-driven automated solver generation shows promise, current paradigms face three critical issues when tackling expensive optimization: factual hallucinations due to deficient domain knowledge, the frequent dismantling of previously established locally optimal structures during refinement, and the prohibitive evaluation costs alongside restricted generalization caused by executing on training instances. To address these issues, we introduce AutoSG, a fully automated workflow directly translating natural language prompts into executable customized solvers. AutoSG features three core innovations: a retrieval-augmented solver generation module strictly grounding code in verified literature; a one-step self-refinement operator introducing task-specific improvements while preserving critical structural components; and an instance-free Elo-based LLM-as-a-Judge evaluation mechanism rapidly establishing global rankings. Extensive evaluations across diverse expensive optimization tasks confirm AutoSG significantly outperforms human-designed state-of-the-art frameworks and existing LLM-generated solvers.