EvoLib：将经验转化为进化知识，实现测试时学习

AI 摘要

微软研究院发布了 EvoLib 框架，使大语言模型能在推理时从自身经验中学习，无需标签或外部反馈。EvoLib 将过往尝试转化为可复用的技能和反思性见解，并通过整合、加权等机制持续优化，使知识随时间变得更通用。在数学推理、代码生成和长程决策任务中，EvoLib 优于现有记忆方法，且对任务顺序具有鲁棒性。该框架无需更新模型，可应用于任何黑盒语言模型。

正文节选

At a glance - Self-supervised. EvoLib enables large language models to learn from their own experience during inference, without requiring ground-truth labels or external feedback. - From experience to knowledge. EvoLib transforms past attempts into reusable skills and reflective insights that can be applied to future tasks. - Knowledge that evolves. Useful skills and insights are continually refined, consolidated, and reweighted, turning instance-specific observations into increasingly general

发布时间：2026-07-31 00:00

抓取时间：2026-08-02 00:26

来源机构：Microsoft Research