COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization

2026-09-10

AI Summary

本文提出COBRA-Skills框架，通过结合上下文引导的优先级策略和基于证据的技能进化方法，有效降低了大型语言模型技能优化的成本和所需数据量。

Abstract

Large language model (LLM) agents can benefit from reusable skills distilled from prior task experience, yet existing skill optimization methods often rely on costly execution-based evaluation and substantial task data. We introduce \textbf{COBRA-Skills}, an efficient framework that formulates skill optimization as budgeted sequential optimization over a dynamically evolving candidate space. COBRA-Skills couples contextual-bandit-guided prioritization with evidence-grounded skill evolution, selectively allocating evaluations to promising or informative candidates while continually refining the skill population from execution feedback. Across six heterogeneous agent benchmarks and three target models, COBRA-Skills consistently achieves the strongest average performance among compared methods, while reducing optimization cost by 55--58\% relative to SkillOpt and using only 50 unique optimization examples per benchmark. Further analyses show that COBRA-Skills remains robust to changes in the agent harness and performs effectively when the target model itself is used for skill generation and refinement.

Problem

• Large language model
• skill optimization
• execution-based evaluation
• task data

Innovation

• contextual bandit
• skill evolution
• budgeted sequential optimization
• execution feedback