发表机构
MBZUAI; McGill University; MirrorSpace Technology; Simon Fraser University（穆罕默德·本·扎耶德人工智能大学; 麦吉尔大学; 镜像空间科技; 西蒙菲莎大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究诊断代码级自主研究循环中的算法模式崩溃问题，提出DAPS方法缓解，有效降低语义集群衰减、提升相对忠实度且保留优化速度。
AI 中文摘要
代码级自主研究循环（ARLs）是自动化机器学习研究中新近出现的具体研究对象。在这类循环中，大型语言模型（LLM）智能体提出对实验训练流水线的修改，执行修改后的流水线，并保留能提升可验证循环内指标的编辑。尽管可执行指标似乎能提供可靠的进展信号，但目前尚不清楚重复的指标驱动型代码编辑是否会带来超出循环范围的泛化性真实改进。我们对该问题进行了系统诊断。在多种实验设置中，我们识别出一种被称为“算法模式崩溃”的鲁棒失效模式：在此状态下，表面层面的编辑多样性保持稳定，但语义与机制层面的多样性崩溃——智能体继续编辑不同代码行，却反复提出相同类型的算法修改。这种崩溃伴随循环内指标增益与独立保留评估测得的增益之间的差距扩大。随后我们提出了Diversity-Aware Proposal Sampling（DAPS），这是一种轻量级缓解方法，结合了类别覆盖重加权、持久编辑记忆和验证门。在将循环内指标、验证门读取的审计指标与循环组件从未访问的盲指标分离的三层协议下，DAPS将编辑的语义集群衰减降低了69.1%，并在盲指标上提升了83.7%的相对忠实度，在审计指标上提升了81.6%，同时保留了循环内优化速度。我们在GitHub上提供了代码仓库。
英文摘要
Code-level autonomous research loops (ARLs) have recently emerged as a concrete object of study in automated machine learning research. In such loops, an LLM agent proposes modifications to an experimental training pipeline, executes the modified pipeline, and retains edits that improve a verifiable in-loop metric. Although executable metrics may appear to provide a reliable signal of progress, it remains unclear whether repeated metric-driven code editing leads to genuine improvements that generalize beyond the loop. We provide a systematic diagnosis of this question. Across various experiment settings, we identify a robust failure mode that we call \textbf{algorithmic mode collapse}. In this regime, surface-level edit diversity remains stable, but semantic and mechanism-level diversity collapse: the agent continues to edit different lines of code while repeatedly proposing the same kinds of algorithmic changes. This collapse is accompanied by a widening gap between in-loop metric gains and gains measured on independent held-out evaluations. We then propose Diversity-Aware Proposal Sampling (\textsc{DAPS}), a lightweight mitigation that combines category-coverage reweighting, persistent edit memory, and a validation gate. Under a three-tier protocol separating the in-loop metric, the audit metric read by the gate, and a blind metric no loop component ever accesses, \textsc{DAPS} reduces semantic-cluster decay of edits by $69.1\%$ and improves relative faithfulness by $83.7\%$ blind and $81.6\%$ audited, while preserving in-loop optimization speed. We provide the code in Github \href{https://github.com/BokwaiHo/arl-mode-collapse}{repository}.
CommentsAccepted by EMNLP 2026
