arXiv：cs.LG· Jackie Baek·· 10 小时前
LLM 能否设计近最优的运筹学算法
LLMs Can Design Near-Optimal OR Algorithms
AI 导读
研究测试了 LLM 为库存控制、排队网络控制和选品优化设计算法的能力。在最难的设置下，模型仅接收问题类别描述和参数范围，返回的固定算法在几乎所有评测实例上匹配或超越原论文最佳方法，最强模型 gpt-5.6-sol 尤其突出。相隔不到八个月发布的模型间性能大幅提升，表明该能力进步迅速。
正文
Abstract:We ask whether large language models (LLMs) can design effective algorithms for well-specified operations research (OR) problems. We study inventory control, queueing network control, and assortment optimization, using benchmark instances from recent papers in each area. We evaluate two levels of LLM use: at level 1, the model receives one problem instance and returns a solution for that instance; at level 2, it receives only the problem class description and broad parameter ranges, and returns an algorithm that maps instance parameters to solutions. Human input is minimal: we give one untuned prompt that describes the problem, and the model has access to a Python sandbox tool with a fixed compute budget.
The strongest model we test, gpt-5.6-sol, matches or outperforms the best method reported in those papers on almost all evaluated instances. This holds even at level 2, where the returned algorithm is fixed before seeing the evaluation instances. Performance also improves sharply across models released less than eight months apart, suggesting that this capability is moving quickly. Thus, for the well-specified operations problems we study, a single untuned LLM query can already produce algorithms competitive with specialized methods. These results suggest that frontier LLMs can be a serious empirical baseline for algorithm design in well-specified OR problems.
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2608.27296 [cs.AI] |
| (or arXiv:2608.27296v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2608.27296 arXiv-issued DOI via DataCite |
Submission history
From: Jackie Baek [view email]
[v1]
Thu, 27 Aug 2026 16:01:11 UTC (108 KB)
[v2]
Tue, 8 Sep 2026 14:35:07 UTC (108 KB)
[v3]
Tue, 13 Oct 2026 21:11:02 UTC (134 KB)
来源：arXiv：cs.LG · arxiv.org