arXiv：cs.AI· Yuwei Gu, Yaoxin Wu, Tong Guo, Wen Song, Zhiguang Cao·· 4 小时前AI 评分46
AutoMIP：面向混合整数线性与非线性规划的自动研究智能体技能
Autoresearch in Mixed-Integer Linear and Nonlinear Programming
AI 导读
AutoMIP 是一个可复用的智能体技能，通过想法池与算法树搜索组织混合整数规划中的长周期自动研究。在 MIPLib 上，它为 60 个实例中的 31 个找到新的最优解；在 MINLPLib 上则为 60 个实例中的 52 个找到新的最优解，在评测的自动研究框架中最终成功率最高。消融实验显示，想法池与算法树搜索对长周期自动研究各有互补贡献。
正文
Abstract:Despite recent progress in autoresearch, applying it to practical operations research problems, typically formulated as NP-hard mixed-integer linear or nonlinear programs (MILPs or MINLPs), remains challenging because effective research requires systematically managing competing ideas and long-horizon experimental trajectories. We introduce AutoMIP, a reusable agent skill for organizing long-horizon autoresearch in mixed-integer programming through idea pooling and algorithm tree search. AutoMIP maintains a persistent pool of complementary candidate ideas while organizing executable experiments into an algorithm tree, enabling the agent to preserve unexplored hypotheses, refine promising algorithms, and switch to alternative methodological directions based on historical states. On MILP and MINLP benchmark cohorts, AutoMIP achieves the highest final success rates among the evaluated autoresearch frameworks. On MIPLib, AutoMIP discovers new best solutions for 31 of 60 instances, surpassing existing autoresearch frameworks. On MINLPLib, it achieves new best solutions for 52 of 60 instances. Ablation studies further demonstrate the complementary contributions of idea pooling and algorithm tree search, highlighting the importance of jointly maintaining diverse research ideas and structured experimental trajectories for long-horizon autoresearch.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2609.39360 [cs.AI] |
| (or arXiv:2609.39360v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.39360 arXiv-issued DOI via DataCite |
Submission history
From: Yaoxin Wu [view email]
[v1]
Wed, 30 Sep 2026 09:18:01 UTC (2,912 KB)
[v2]
Thu, 1 Oct 2026 21:03:56 UTC (2,912 KB)
来源：arXiv：cs.AI · arxiv.org