SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models

  • 类型:arxiv
  • 标识:2608.10538
  • 链接:https://arxiv.org/abs/2608.10538
  • 主分类:agent
  • 形态:application
  • 被引:0
  • 被引来源:Semantic Scholar
  • S2被引:0
  • 影响力被引:0
  • TLDR:SKILLER is a natural-language-driven reinforcement learning framework designed to automatically generate executor-specific skills for small models, which employs a strong model as the actor and critic, treats the small-model agent system as the environment, and propagates all reinforcement learning signals entirely via natural language.
  • 副分类:llm-infra
  • 待LLM分类:否
  • 标题中文:SKILLER:面向小型语言模型可复用技能提取的语言级强化学习
  • TLDR中文:SKILLER 是一个由自然语言驱动的强化学习框架,旨在为小模型自动生成执行器特定的 skills,使用强模型作为 actor 和 critic,将小模型 Agent 系统视为环境,并通过自然语言完全传递所有强化学习信号。
  • 来源文件
  • /inbox/tom/_candidates/2026-08-14-agent-rag-longcontext-candidates.json
  • [S2 enrich]