arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~

SkillContrast:用于智能体技能重排序的差异引导文本选择

SkillContrast: Difference-Guided Text Selection for Agent Skill Reranking

Jiandong Ding, Honglei Ji, Ming Liu, Tao Duan

arXiv 2610.11650首次发表:更新:

发表机构

Huawei Technologies Co., Ltd.; Tongji University School of Medicine(华为技术有限公司; 同济大学医学院)

机构由 AI 辅助整理,请以论文原文为准。

AI 中文总结

SkillContrast是一种无需训练的文本选择器,通过保留检索技能的差异文本,在SameCapRisk-Bench上相比TF-IDF查询选择提升了智能体技能重排序的干净命中数,同时减少了模型输入标记。

AI 中文摘要

相似的智能体技能可能共享指令,但使用条件存在差异。基于查询的文本选择可能会保留共享指令,却忽略这些差异。我们提出SkillContrast,这是一种无需训练的选择器,可比较检索到的技能,为预训练的重排序器保留带有局部上下文的差异文本。在来自SameCapRisk-Bench的1235个请求上,在相同的每个候选输入长度下,它比TF-IDF查询选择多产生54-72个干净命中(即检索到有用技能且无其标记的风险同类技能的请求),且在2种检索器和2种重排序器规模下均如此。长度匹配的组件替换表明,在主要设置中,差异文本是主要贡献因素,而较小的混合上下文效应也存在。与完整技能文本相比,SkillContrast使用的模型输入标记减少了51.1-58.8%,在0.6B规模下干净命中数少10-18个,在4B规模下干净命中数匹配或更高。因此,相对于完整技能的差异补充了查询相关性,用于选择紧凑的重排序输入。

英文摘要

Similar agent skills can share instructions but differ in their conditions of use. Query-based text selection may retain shared instructions and omit these distinctions. We introduce SkillContrast, a training-free selector that compares retrieved skills and retains their differing text with local context for a pretrained reranker. On 1,235 requests from SameCapRisk-Bench, it yields 54-72 more clean hits (requests that retrieve a helpful skill without its marked risky sibling) than TF-IDF query selection at identical per-candidate input lengths, across 2 retrievers and 2 reranker sizes. Length-matched component replacements identify differing text as the main contributor in the primary setting, with smaller, mixed context effects. Relative to full skill bodies, SkillContrast uses 51.1-58.8% fewer model-input tokens, with 10-18 fewer clean hits at 0.6B and matching or higher observed clean-hit counts at 4B. Candidate-relative differences thus complement query relevance in selecting compact reranking inputs.

Comments5 pages, 2 figures, 3 tables

论文原文

arXiv 摘要页 · PDF 原文 · HTML 原文

↑