---
id: 20260816-T0-03
title: "ε-MemEvo：让LLM代码进化系统跨任务复用搜索经验"
title_en: "ε-MemEvo: Cross-Task Memory Transfer Boosts LLM Program Evolution Efficiency"
url: https://ai.daily.yangsir.net/daily/20260816-T0-03
issue_date: 2026-08-16
publish_date: 2026-08-15T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2608.12522
---

# ε-MemEvo：让LLM代码进化系统跨任务复用搜索经验

arXiv新论文介绍ε-MemEvo，一种针对LLM程序进化系统的跨任务记忆转移方法。像FunSearch和AlphaEvolve这类系统虽能发现新算法，但通常每个任务独立优化，搜索经验在完成后被丢弃。ε-MemEvo将过去任务的经验通过自适应机制转移到新任务中，加速进化过程，减少重复搜索的计算浪费。

## English Version

**ε-MemEvo: Cross-Task Memory Transfer Boosts LLM Program Evolution Efficiency**

A new arXiv paper presents ε-MemEvo, an adaptive cross-task memory transfer method for LLM-based program evolution systems like FunSearch and AlphaEvolve. These systems typically optimize each task in isolation, discarding search experience. ε-MemEvo transfers knowledge from prior tasks to new ones, accelerating evolution and cutting redundant computation.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2608.12522)

**详情页**：https://ai.daily.yangsir.net/daily/20260816-T0-03

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*