大型语言模型 (LLM) 代理越来越多地应用于长期任务，例如科学发现和机器学习工程 (MLE),，其中持续的自我进化成为关键能力。然而,现有的MLE代理遭受分支间信息隔离,无记忆搜索,和缺乏分层控制,的困扰，这些都阻碍了长期优化。我们提出 MLEvolve, 是一个基于 LLM 的自进化多代理框架，用于端到端机器学习算法发现。通过将树搜索扩展到渐进式 MCGS,，MLEvolve 可以通过基于图的参考边实现跨分支信息流，并通过受熵启发的渐进式计划逐渐将搜索从广泛探索转变为集中利用。为了让代理随着积累的经验, 不断发展，我们引入了回顾性记忆,，它将冷启动领域知识库与动态全局记忆相结合，用于特定任务的经验检索和重用。为了稳定的长范围迭代,，我们利用自适应编码模式进一步将战略规划与代码生成分离。 MLE-Bench 的评估表明，MLEvolve 在多个维度上实现了最先进的性能，包括在 12 小时预算(（标准运行时间的一半)）下的平均奖牌率和有效提交率。此外，, MLEvolve 在数学算法优化任务上也优于包括 AlphaEvolve 在内的专业算法发现方法,，展示了强大的跨域泛化能力。我们的代码可以在这个 https URL 上找到。

Large language model (LLM) agents are increasingly applied to long-horizon tasks such as scientific discovery and machine learning engineering (MLE), where sustained self-evolution becomes a key capability. However, existing MLE agents suffer from inter-branch information isolation, memoryless search, and lack of hierarchical control, which together hinder long-horizon optimization. We present MLEvolve, an LLM-based self-evolving multi-agent framework for end-to-end machine learning algorithm discovery. By extending tree search to Progressive MCGS, MLEvolve enables cross-branch information flow through graph-based reference edges and gradually shifts the search from broad exploration to focused exploitation with an entropy-inspired progressive schedule. To allow the agent to evolve with accumulated experience, we introduce Retrospective Memory, which combines a cold-start domain knowledge base with a dynamic global memory for task-specific experience retrieval and reuse. For stable long-horizon iteration, we further decouple strategic planning from code generation with adaptive coding modes. Evaluation on MLE-Bench shows that MLEvolve achieves state-of-the-art performance across multiple dimensions including average medal rate and valid submission rate under a 12-hour budget (half the standard runtime). Moreover, MLEvolve also outperforms specialized algorithm discovery methods including AlphaEvolve on mathematical algorithm optimization tasks, demonstrating strong cross-domain generalization. Our code is available at this https URL.

科目: 人工智能 (cs.AI); 计算和语言 (cs.CL)

Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)