AIMay 9

SearchSkill: Teaching LLMs to Use Search Tools with Evolving Skill Banks

arXiv:2605.0903886.3Has Code
Predicted impact top 26% in AI · last 90 daysOriginality Incremental advance
AI Analysis

For open-domain question answering, this provides a lightweight alternative to treating search as an undifferentiated action, improving query quality and answer accuracy.

SearchSkill teaches LLMs to issue better search queries by making query planning explicit through reusable, evolving search skills, improving exact match on knowledge-intensive QA benchmarks and retrieval behavior with fewer copied queries and more correct answers within a small search budget.

Teaching language models to use search tools is not only a question of whether they search, but also of whether they issue good queries. This is especially important in open-domain question answering, where broad or copied queries often waste retrieval budget and derail later reasoning. We propose \Ours, a framework that makes query planning explicit through reusable search skills. At each step, the model first selects a skill, then generates a search or answer action conditioned on the selected skill card. The skill inventory itself is not fixed: SearchSkill maintains an evolving SkillBank, expands or refines it from recurrent failure patterns, and reconstructs affected trajectories before supervised training. The resulting two-stage SFT recipe aligns training with the inference-time protocol of skill selection followed by skill-grounded execution. Across open-source and closed-source models, SearchSkill improves exact match on knowledge-intensive QA benchmarks and yields better retrieval behavior, including fewer copied first queries, more atomic hop-focused queries, and more correct answers within a small search budget. These results suggest that explicit skill-conditioned query planning is a lightweight alternative to treating search as an undifferentiated action.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes