arXiv cs.AI· Wenhan Xiao, Ziwei Zhang, Xingcheng Fu, Chuanyue Yu, Haonan Yuan, Qingyun Sun, Runhua Xu, Jianxin Li·· 2 小时前AI 评分38
SKALOS：基于能力图学习区分相似技能
SKALOS: Learning to Distinguish Similar Skills with Capability Graphs
AI 导读
arXiv 论文提出 SKALOS，一种图原生重排序器，通过显式能力证据比较语义相似的技能，并用 semiring 能力 witness 学习按最弱证据给分支打分。在 SkillRet 上，SKALOS 的 NDCG@10 比匹配的纯文本重排序器提升 2.68 个百分点；witness 引导的技能选择还将 ALFWorld 任务成功率从 22% 提升至 29%。
正文 · 原文
Abstract:Skills provide LLM agents with reusable procedural knowledge, making appropriate skill selection essential for reliable agent behavior. As skill libraries expand, distinguishing semantically similar candidates becomes increasingly difficult. Existing methods encode or rewrite skill descriptions for relevance-based ranking, yet heterogeneous descriptions can obscure capability differences. To address this challenge, we propose SKALOS, a graph-native reranker that compares skills via explicit capability evidence. Its multi-hierarchy graph aligns equivalent expressions across skills while preserving skill-specific capability evidence. Since individually relevant capability elements may not jointly support a query, semiring-based capability witness learning scores each branch by its weakest evidence, checks joint support for the extracted requirements within one implementation, and selects the strongest alternative. To reduce the influence of extraction errors, we develop ranking-feedback graph calibration that adjusts unreliable evidence for candidate comparison. Experiments on SkillRet show that SKALOS improves NDCG@10 by 2.68 percentage points over a matched text-only reranker. Witness-guided skill selection also increases ALFWorld task success rate from 22% to 29%.
| Comments: | 12 pages, 7 figures, 19 tables |
| Subjects: | Artificial Intelligence (cs.AI) |
| ACM classes: | H.3.3; I.2.6 |
| Cite as: | arXiv:2610.34728 [cs.AI] |
| (or arXiv:2610.34728v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34728 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Wenhan Xiao [view email]
[v1]
Mon, 26 Oct 2026 02:25:42 UTC (937 KB)
来源：arXiv cs.AI · arxiv.org