arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2604.16804cs.LGcs.AI
收藏

AutoOR: 一种可扩展的后训练语言模型用于自动形式化运筹学问题

AutoOR: Scalably Post-training LLMs to Autoformulate Operations Research Problems

The Moonshot Factory(谷歌登月工厂);University of Oxford(牛津大学)

Sumeet Ramesh Motwani, Chuan Du, Aleksander Petrov, Christopher Davis, Philip Torr, Antonio Papania-Davis, Weishi Yan

AI总结:

AutoOR通过合成数据生成与强化学习管道,训练语言模型自动形式化运筹学问题,实现线性、混合整数和非线性优化问题的高效解决,取得六个经典运筹学基准测试的领先或竞争性结果。

摘要:

优化问题在制造、物流、调度等工业领域决策中至关重要。将复杂描述转化为求解器准备的公式需要专门的运筹学(OR)专业知识,难以扩展。我们提出了AutoOR,一种可扩展的合成数据生成和强化学习管道,训练语言模型在自然语言中自动形式化跨线性、混合整数和非线性类别的优化问题。AutoOR从标准优化形式生成验证训练数据,并使用求解器执行反馈作为RL后训练的奖励信号。在8B模型上应用AutoOR,在六个经典OR基准测试中取得最先进的或竞争性结果,显著匹配更大的前沿模型。对于涉及物理动态的非线性问题类别,前沿模型得分接近0%,我们引入了课程RL策略,从有限的初始训练数据出发,使该类别在后训练中可处理。我们相信,此类方法如AutoOR可以显著加速工业决策中的AI应用。

英文摘要:

Optimization problems are central to decision-making in manufacturing, logistics, scheduling, and other industrial settings. Translating complicated descriptions of these problems into solver-ready formulations requires specialized operations research (OR) expertise, making it hard to scale. We present AutoOR, a scalable synthetic data generation and reinforcement learning pipeline that trains LLMs to autoformulate optimization problems specified in natural language across linear, mixed-integer, and non-linear categories. AutoOR generates verified training data from standard optimization forms and uses solver execution feedback as the reward signal for RL post-training. AutoOR applied to an 8B model achieves state-of-the-art or competitive results across six established OR benchmarks, matching significantly larger frontier models. For a non-linear problem class involving physical dynamics, where frontier models score near 0%, we introduce a curriculum RL strategy that bootstraps from limited initial training data to make this class tractable for post-training. We believe that methods such as AutoOR can significantly accelerate industrial decision-making with AI.

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。

↑