COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization
COBRA-Skills uses contextual bandits to guide LLM agent skill evolution, cutting optimization cost 55-58% versus SkillOpt while topping six agent benchmarks.
COBRA-Skills formulates LLM agent skill optimization as budgeted sequential optimization over a dynamically evolving candidate space. It couples contextual-bandit-guided prioritization with evidence-grounded skill evolution, selectively spending execution-based evaluations on promising candidates while refining skills from feedback. Across six heterogeneous agent benchmarks and three target models, it achieves the strongest average performance while reducing optimization cost by 55-58% relative to SkillOpt using only 50 unique optimization examples per benchmark. The method remains robust to agent harness changes and works when the target model generates its own skills.
- Frames skill optimization as budgeted sequential optimization over evolving candidates
- Contextual bandits prioritize which skill candidates receive costly evaluations
- 55-58% cost reduction versus SkillOpt with only 50 examples per benchmark
- Best average performance across six benchmarks and three target models
From the original article
1 min read at huggingface.co
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 COBRA-Skills, an efficient framework that formulates skill optimization as budgeted sequential optimization over a dynamically evolving candidate space.
By Pingchen Lu, Xiangyi Wang, Xiang Li · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.