arXiv cs.CL· Seockbean Song, Youngsik Yoon, Sungjae Lee, Siwei Wang, Wei Chen, Jungseul Ok·· 11 小时前AI 评分26
HoliDivEvolve：LLM 引导进化搜索中的整体多样性管理
Holistic Diversity Management in LLM-Guided Evolutionary Search
AI 导读
针对 LLM 引导进化搜索中岛屿模型难以保证搜索方向差异的问题，研究者提出 HoliDivEvolve，从初始化、岛屿轨迹到跨岛迁移全流程管理多样性。该方法引入多样化初始化、按当前得分、近期增长与剩余预算分配算力的轨迹自适应岛屿选择，以及增长自适应迁移机制。多基准实验显示其在多数任务上取得有竞争力的表现。
正文 · 原文
Abstract:LLM-guided evolutionary search iteratively selects existing programs, uses LLMs as mutation operators to generate a new program, and evaluates it to guide subsequent search. To preserve diversity and reduce the risk of premature convergence, it often adopts the island model, in which groups of programs, called islands, evolve separately. However, this structure alone does not guarantee distinct search directions, as islands may start from similar programs or converge toward similar solutions during evolution. This suggests that diversity must be considered throughout the search process, from initializing islands with distinct programs to developing island trajectories and sharing discoveries across islands. To this end, we propose HoliDivEvolve , a holistic approach to establishing and effectively utilizing island diversity throughout evolutionary search. Specifically, we introduce diverse initialization to establish different initial programs, trajectory adaptive island selection to allocate computation based on current score, recent growth, and the remaining search budget, and growth adaptive migration to share alternative ideas across islands. Experiments across diverse benchmarks show that HoliDivEvolve achieves competitive performance on most evaluated tasks. We further conduct controlled analyses to examine how our design choices affect search behavior and final performance.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.38407 [cs.CL] |
| (or arXiv:2610.38407v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.38407 arXiv-issued DOI via DataCite |
Submission history
From: Seockbean Song [view email]
[v1]
Wed, 28 Oct 2026 07:52:13 UTC (2,152 KB)
[v2]
Fri, 30 Oct 2026 06:18:25 UTC (2,152 KB)
来源：arXiv cs.CL · arxiv.org