arXiv cs.AI· Changye Li, Zijian Zhang, Muheng Li, Yiduo Pang, Tingjun Wu, Menghan Li, Yitong Liu, Hongkai Guo, Mengchi Chen, Chenyu He, Kaisheng Ma, Wei Xu, Song Han, Sihao Liu, Ligeng Zhu·· 6 天前AI 评分36
Humanize MLE Agents（HMA）：多智能体会话交替突破单智能体极限
Humanize MLE Agents: Multi-Agent Session Alternation Beyond Single-Agent Limits
AI 导读
针对长周期 MLE 任务中单会话推理强度下降的"长周期减速"与定制 harness 收益随模型升级消失的"harness 陈旧"问题，研究者提出 Humanize MLE Agents（HMA），让两个智能体在共享工作区交替执行、每会话最多 k 次实验并重置上下文。
正文 · 原文
Authors:Changye Li, Zijian Zhang, Muheng Li, Yiduo Pang, Tingjun Wu, Menghan Li, Yitong Liu, Hongkai Guo, Mengchi Chen, Chenyu He, Kaisheng Ma, Wei Xu, Song Han, Sihao Liu, Ligeng Zhu
Abstract:Existing LLM-based agent approaches for long-horizon machine learning engineering (MLE) often use context engineering, implemented through custom harnesses that manage tools, skills, and accumulated experience. However, we find that as single-session execution lengthens, the agent's reasoning intensity declines alongside slowing progress, summarized as long-horizon deceleration, and the benefits of these custom harnesses vanish as the underlying model improves, summarized as harness staleness. These findings suggest that their benefits may not persist across versions, which motivates a workflow-level approach in which agents build on one another's work across sessions. We model these alternating agent sessions as a context-reset semi-Markov decision process (CR-SMDP) and establish sufficient conditions for alternation to exceed single-agent reference limits, derive an explicit experiment-budget bound, and characterize when session caps preserve this advantage. Guided by this analysis, we introduce Humanize MLE Agents (HMA), which alternates two agents on a shared workspace, limits each session to at most k experiments and resets the context between sessions, without additional task-specific harness or prior knowledge supplements. Across 75 Kaggle tasks in MLE-bench, HMA reaches a state-of-the-art any-medal rate of 78.2%, exceeding the strongest open-source baseline by 8.0 points with a quarter of its time budget. Moreover, among six configurations spanning five heterogeneous agent pairs, HMA improves any-medal rate over the constituent baselines by 6.4 points on average, demonstrating cross-pair generality. A supplementary case on kernel optimization reaches 878 cycles, within 1.6% of the community leaderboard leader, demonstrating applicability beyond MLE.
| Comments: | 32 pages, 10 figures. Code: this https URL |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.13494 [cs.AI] |
| (or arXiv:2610.13494v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.13494 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Changye Li Mr. [view email]
[v1]
Fri, 9 Oct 2026 12:26:34 UTC (686 KB)
来源：arXiv cs.AI · arxiv.org