让弱模型通过老师经验学习，无需训练就能高效执行任务。
AgentBrew: Lifelong Knowledge Brewing from Strong Teachers to Weak LLM Agents
- 用失败触发机制将老师经验转为可执行笔记
- 在编码、数学等任务上超越基线模型性能
- 适合资源有限但需持续学习的部署场景
部署大型语言模型（LLM）代理通常需要一个轻量级的测试时学生模型，即使训练时存在更强的教师模型。本文研究知识酿造：将教师的交互经验提炼为学生持久的外部记忆。关键在于无需权重更新、专家示范、真实标签或测试时教师访问。这一设定面临两大挑战：环境仅提供稀疏的二值反馈，且教师撰写的笔记必须针对明显更弱的学生具备可执行性。为此，我们提出AgentBrew，包含两个耦合组件：首先，失败触发式教师——Ralph Loop通过将学生失败转化为经环境验证的笔记来缓解稀疏反馈；其次，学生感知的合成机制校准教师知识至弱执行者的操作粒度，生成模型专属、可操作的指导。在编码、数学和工具使用任务上的广泛评估与全面消融实验表明，这种非对称、无训练的知识酿造范式能生成高度能力且可部署的LLM代理。
原文摘要 · Abstract (English)
Deploying LLM agents typically requires a compact test-time student, even if a stronger teacher is available during training. We study knowledge brewing: distilling a teacher's interactive experience into a persistent external memory for the student. Crucially, this requires no weight updates, expert demonstrations, ground-truth labels, or test-time teacher access. This setting poses two challenges: environments provide only sparse, binary feedback, and teacher-authored notes must be inherently tailored to be concretely executable by a substantially weaker student. To address these hurdles, we propose AgentBrew, comprising two coupled components. First, a failure-triggered teacher--Ralph Loop mitigates sparse feedback by transforming student failures into environment-validated notes. Second, student-aware synthesis calibrates teacher knowledge to the weak executor's operational granularity, yielding model-specific, actionable guidance. Extensive evaluations and comprehensive ablations across coding, math, and tool-use tasks demonstrate that this asymmetric, training-free brewing paradigm produces highly capable yet deployable LLM agents.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。