arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2604.18607cs.NEcs.AI
收藏

TurboEvolve:迈向快速且稳健的LLM驱动程序进化

TurboEvolve: Towards Fast and Robust LLM-Driven Program Evolution

Artificial Intelligence Thrust, Hong Kong University of Science and Technology (Guangzhou)(香港科学与技术大学(广州)人工智能方向)

Yang Yang, Zining Zhong, Jindong Li, Jiemin Wu, Kaishen Yuan, Wenshuo Chen, Menglin Yang, Yutao Yue

AI总结:

TurboEvolve通过多岛屿进化框架提升样本效率和鲁棒性,在固定评估预算下实现更可靠的程序进化进展。

摘要:

LLM驱动的程序进化可以发现高质量的程序,但其成本和运行间方差阻碍了可靠进步。我们提出了TurboEvolve,一种多岛屿进化框架,通过引入 verbalized Sampling 和在线调度器,在固定评估预算下提高样本效率和鲁棒性。受进化算法中的多子策略启发,TurboEvolve 引入 verbalized Sampling,促使 LLM 生成 K 个多样候选方案并显式分配采样权重,同时在线调度器根据停滞状态扩展探索并减少稳态进展时的开销。为利用现有解决方案池,我们进一步提出“种子池注入”,通过聚类种子并跨岛屿分配受控扰动和精英保留,平衡多样性和精细化。在多个程序优化基准测试中,TurboEvolve 在较低预算下 consistently 实现更强性能,并在多个任务上改进最佳已知解决方案。

英文摘要:

LLM-driven program evolution can discover high-quality programs, but high API cost and run-to-run variance hinder reliable progress. We propose TurboEvolve, a multi-island evolutionary framework that improves sample efficiency and stability under matched API-cost budgets. Inspired by multiple-offspring strategies in evolutionary algorithms, TurboEvolve introduces Verbalized Sampling, which elicits K diverse candidates in one LLM call while using self-assigned probabilities only as generation signals, together with an online scheduler that adapts K to expand exploration under stagnation and reduce redundant generation during steady progress. To exploit existing solution pools, TurboEvolve further introduces island-aware initialization, which clusters seed programs into mode-specific subpopulations and applies controlled cross-cluster mixing with elitist preservation to balance diversity, sharing, and refinement. We also curate a reusable cross-task solution-pool dataset to support warm-start initialization studies and matched-resource comparisons. Across mathematical discovery and GPU-kernel optimization tasks, TurboEvolve achieves stronger cost-quality trade-offs and more stable performance than controlled baselines, while reaching competitive reported scores.

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。

↑