SkillContrast: Difference-Guided Text Selection for Agent Skill Reranking
We introduce SkillContrast, a training-free selector that compares retrieved skills and retains their differing text with local context for a pretrained reranker.
ProofPaper ↗
Key points
- 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.
- 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.
- Candidate-relative differences thus complement query relevance in selecting compact reranking inputs.
Sources (1)
- [1]SkillContrast: Difference-Guided Text Selection for Agent Skill RerankingarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 10:25 AM
We introduce SkillContrast, a training-free selector that compares retrieved skills and retains their differing text with local context for a pretrained reranker.
Similar agent skills can share instructions but differ in their conditions of use.
Extractive summary: sentences quoted from the sources.