SEDIMA：进化搜索智能体的跨运行分层洞察记忆

中文摘要 AI 辅助

大语言模型（LLM）驱动的进化搜索是自动化程序和算法发现的一种强大范式，然而现有系统在很大程度上是无记忆的：每次运行都从零开始探索，因此智能体会反复重新发现相同的改进，并再次遇到相同的死胡同。我们引入了SEDIMA，一种用于进化搜索智能体的持久化分层洞察记忆。SEDIMA将原始轨迹提炼为自然语言洞察，使用注意力加权质心按语义相似性进行聚类，并检索相关指导以调节未来的变异，从而在多次运行和问题之间积累可迁移的知识，而非局限于单一轨迹内。作为一个无需修改搜索算子的即插即用模块，在固定预算为100个评估候选的情况下，SEDIMA在AlgoTune上将平均最终性能提升了5.5%，在ALE-Bench LITE上提升了6.6%。在OpenEvolve下，SEDIMA在五个评估骨干网络上平均需要减少32.3%的迭代次数即可达到基线最佳性能。

英文摘要

Large language model (LLM)-driven evolutionary search is a powerful paradigm for automated program and algorithm discovery, yet existing systems are largely memoryless: each run explores from scratch, so agents repeatedly rediscover the same improvements and re-encounter the same dead ends. We introduce SEDIMA, a persistent hierarchical insight memory for evolutionary search agents. SEDIMA distills raw traces into natural-language insights, clusters them by semantic similarity using attention-weighted centroids, and retrieves relevant guidance to condition future mutations, accumulating transferable knowledge across runs and problems rather than within a single trajectory. As a drop-in module that leaves the search operators unmodified, SEDIMA improves average final performance by 5.5% on AlgoTune and 6.6% on ALE-Bench LITE under a fixed budget of 100 evaluated candidates. Under OpenEvolve, SEDIMA requires 32.3% fewer iterations on average to reach baseline-best performance across the five evaluated backbones.

发表机构

• University of British Columbia（不列颠哥伦比亚大学）
• Huawei Technologies Canada（华为加拿大技术公司）

机构由 AI 辅助整理，请以论文原文为准。