Optimal Skill Selection for LLM Agents with Provable Bicriteria Guarantees

2026-08-20

AI Summary

本文针对LLM代理技能选择问题，提出了一种名为BPS的优化算法，在保证性能的同时减少了令牌使用量。

Abstract

Loading reusable skill documents into a bounded context window is now the primary way large language model (LLM) agents acquire task-specific capabilities, which makes skill selection a first-order determinant of task performance and token cost. Yet current agents score skills independently by semantic relevance and assemble the set by top-$k$ or greedy packing, with no quality guarantee or cost awareness on the selected set. As a result, redundant or poorly chosen skills waste scarce context tokens and can even degrade performance. We give the first model of how the selected skill set shapes execution outcomes and cast skill selection as an optimization problem: choose a skill set under a hard token budget to maximize a monotone submodular benefit minus context penalty. For this problem, we develop Best Prefix Selection (BPS), a polynomial-time algorithm, and prove, to our knowledge, the first performance guarantee for skill selection: a bicriteria $(1-1/e,1)$ approximation whose benefit coefficient is optimal in polynomial time. On a contamination-controlled BigCodeBench variant, BPS outperforms all the baselines, reaching $0.73$ measured task success versus $0.20$--$0.52$ for released skill routers, text retrievers, and the executor's own selection, on $28\%$ fewer tokens than the strongest released router.

Problem

• Skill Selection
• Large Language Models
• Token Budget
• Performance Guarantee
• Context Window

Innovation

• Best Prefix Selection
• Bicriteria Guarantees
• Submodular Benefit
• Token Budget
• Skill Selection