arXiv：cs.LG· Md Mehrab Tanjim, Jayakumar Subramanian, Branislav Kveton, Xiang Chen, Subhojyoti Mukherjee, Anlan Zhang, Sungchul Kim, Sunav Choudhury, Somdeb Sarkhel·· 16 小时前
MOCHA：面向 Agent 技能优化的多目标 Chebyshev 退火方法
MOCHA: Multi-Objective Chebyshev Annealing for Agent Skill Optimization
AI 导读
针对技能优化中准确性与经济性难以兼顾的问题，研究者提出 MOCHA（Multi-Objective CHebyshev Annealing），通过 Chebyshev 标量化选择技能并接受扩展 Pareto 前沿的改进，从探索退火到利用。
正文
Abstract:LLM agents organize behavior through skills -- structured natural-language specifications that govern how an agent reasons, retrieves, and responds. Unlike monolithic prompts, skills are multi-field artifacts with platform limits: descriptions are indexed for routing, instruction bodies are loaded into a shared context window, and every extra line costs the agent context it could spend elsewhere. Skill optimization is therefore inherently multi-objective: a skill should be accurate and economical at once. Existing prompt optimizers collapse these goals into a single score, and we find that this makes them stall: in most of their runs, 1000 rollouts yield zero progress over the seed skill. We introduce MOCHA (Multi-Objective CHebyshev Annealing), which selects skills to improve by Chebyshev scalarization and accepts improvements that expand the Pareto front, annealing from exploration to exploitation. Chebyshev scalarization scores a skill by its largest weighted shortfall from a perfect skill; with random weights, it can select any skill on the Pareto front for improvement, including trade-offs that an average would never favor. Across six diverse skills and two models, MOCHA escapes the seed in nearly every run, achieving 7.9% higher mean accuracy than the strongest baseline (up to 23.7% on DebugBench with GPT 5.5 and 14.9% on FEVER with Claude Haiku 4.5) while discovering Pareto fronts with the highest hypervolume and nearly twice as many Pareto-optimal skills.
| Comments: | Preprint. 24 pages, 20 figures, 6 tables |
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE) |
| ACM classes: | I.2.7; I.2.6; I.2.4; I.2.8 |
| Cite as: | arXiv:2605.19330 [cs.AI] |
| (or arXiv:2605.19330v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2605.19330 arXiv-issued DOI via DataCite |
Submission history
From: Md Mehrab Tanjim [view email]
[v1]
Tue, 19 May 2026 04:07:41 UTC (656 KB)
[v2]
Wed, 14 Oct 2026 17:35:11 UTC (788 KB)
来源：arXiv：cs.LG · arxiv.org