EvoOptiGraph：基于图结构生成的弱点驱动协同进化优化建模
EvoOptiGraph: Weakness-Driven Coevolution via Graph-Based Structural Generation for Optimization Modeling
AI总结：
提出EvoOptiGraph框架，通过图结构表示混合整数线性规划并应用进化算子生成多样实例，结合监督微调和强化学习实现数据与模型的协同进化，显著提升优化建模的准确性和泛化能力。
摘要：
利用大语言模型从自然语言自动化优化建模面临两个关键挑战。首先，训练语料缺乏结构多样性。其次，数据生成流程保持静态且与模型学习脱钩。为应对这些挑战，我们提出EvoOptiGraph，一个新颖的框架，其中数据和模型由模型弱点驱动协同进化。EvoOptiGraph将每个混合整数线性规划表示为属性二分图，并应用保持有效性的进化算子生成结构多样的实例。进化后的图通过确定性编译和验证回译转换为求解器代码和自然语言。训练分两个阶段：在初始数据集上进行监督微调，随后进行带可验证奖励的强化学习，其中图衍生的弱点信号指导针对模型失败生成新实例。这形成了一个持续更新训练分布的闭环。在六个公开数据集上的实验结果表明，EvoOptiGraph在准确性、可执行性和泛化性上显著优于更大的通用模型、代理方法和专用基线。这些结果表明，针对性的数据-模型协同进化是提升大语言模型在优化建模任务上表现的有效策略。
英文摘要：
Automating optimization modeling from natural language with large language models (LLMs) faces two key challenges. First, training corpora lack structural diversity. Second, data generation pipelines remain static and decoupled from model learning. To address these challenges, we propose EvoOptiGraph, a novel framework where data and model co-evolve, driven by model weaknesses. EvoOptiGraph represents each mixed-integer linear program (MILP) as an attributed bipartite graph and applies validity-preserving evolutionary operators to generate structurally diverse instances. The evolved graphs are converted into solver code and natural language via deterministic compilation and verified back-translation. Training proceeds in two stages: supervised fine-tuning (SFT) on an initial dataset, followed by reinforcement learning with verifiable rewards (RLVR), where graph-derived weakness signals guide the generation of new instances targeting the model's failures. This forms a closed loop that continuously updates the training distribution. Empirical results on six public datasets show that EvoOptiGraph significantly outperforms larger generalist models, agentic methods, and specialized baselines in accuracy, executability, and generalization. These results demonstrate that targeted data-model coevolution is an effective strategy for improving LLMs on optimization modeling tasks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
