MARS：模块化代理与反思搜索用于自动化AI研究
MARS: Modular Agent with Reflective Search for Automated AI Research
AI总结：
本文提出MARS框架，通过预算感知规划、模块化构建和比较反思记忆解决复杂机器学习工程任务中的执行成本与性能归因问题，实现开放源代码框架在MLE-Bench上的最佳性能。
摘要：
自动化AI研究的关键瓶颈在于执行复杂的机器学习工程（MLE）任务。MLE不同于一般软件工程，因其计算成本高昂（例如模型训练）和性能归因不透明。当前基于LLM的代理在此方面表现不佳，常生成忽视执行成本和因果因素的单体脚本。我们引入MARS（模块化代理与反思搜索），一种优化于自主AI研究的框架。MARS依赖三个支柱：（1）通过成本受限的蒙特卡洛树搜索（MCTS）进行预算感知规划，以显式平衡性能与执行成本；（2）模块化构建，采用“设计-分解-实现”流程来管理复杂的研究存储库；（3）比较反思记忆，通过分析解决方案差异来解决信用分配问题，从而提炼出高信号的洞察。MARS在可比条件下，在开放源代码框架中实现了MLE-Bench上的最佳性能，与全球排行榜上顶尖方法竞争性相当。此外，系统表现出定性“啊哈！”时刻，其中所有使用的63%的教训源自跨分支转移，表明代理能有效在搜索路径间泛化洞察。
英文摘要：
A critical bottleneck in automating AI research is the execution of complex machine learning engineering (MLE) tasks. MLE differs from general software engineering due to computationally expensive evaluation (e.g., model training) and opaque performance attribution. Current LLM-based agents struggle here, often generating monolithic scripts that ignore execution costs and causal factors. We introduce MARS (Modular Agent with Reflective Search), a framework optimized for autonomous AI research. MARS relies on three pillars: (1) Budget-Aware Planning via cost-constrained Monte Carlo Tree Search (MCTS) to explicitly balance performance with execution expense; (2) Modular Construction, employing a "Design-Decompose-Implement" pipeline to manage complex research repositories; and (3) Comparative Reflective Memory, which addresses credit assignment by analyzing solution differences to distill high-signal insights. MARS achieves state-of-the-art performance among open-source frameworks on MLE-Bench under comparable settings, maintaining competitiveness with the global leaderboard's top methods. Furthermore, the system exhibits qualitative "Aha!" moments, where 63% of all utilized lessons originate from cross-branch transfer, demonstrating that the agent effectively generalizes insights across search paths.
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。
