---
id: 20260513-T0-10
title: "SkillLens：自适应技能复用降低LLM成本"
title_en: "SkillLens: Adaptive Skill Reuse for Cost-Efficient LLMs"
url: https://ai.daily.yangsir.net/daily/20260513-T0-10
issue_date: 2026-05-13
publish_date: 2026-05-12T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2605.08386
---

# SkillLens：自适应技能复用降低LLM成本

arXiv新论文提出SkillLens方法，解决LLM智能体技能库中相关性与成本矛盾问题。现有方法将技能视为扁平化提示块，而SkillLens实现多粒度自适应复用，在保持性能的同时降低计算成本。实验验证其效率优势。

## English Version

**SkillLens: Adaptive Skill Reuse for Cost-Efficient LLMs**

New arXiv paper proposes SkillLens to address relevance-cost tension in LLM agent skill libraries. Existing systems treat skills as flat prompts, while this method enables adaptive multi-granularity reuse. Reduces computational cost while maintaining performance.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2605.08386)

**详情页**：https://ai.daily.yangsir.net/daily/20260513-T0-10

---

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