CompSkill：把执行经验编译进技能检索，提升智能体任务完成率
CompSkill: Compiling Execution Experience for Skill Retrieval
一个开源的技能检索方案，检索基准 NDCG@10 从 64.0 提到 70.9，ToolQA 上比 BM25 和 RankGPT 多完成不少任务，做智能体的可以看看它的 KeyOpt 思路。
CompSkill 用学习到的 512 维向量索引技能文档，并通过 KeyOpt 把同类任务的历史执行对比编译成技能偏移与受限键残差，只改检索结果、不改文档文本。在 SkillRet 检索基准上，该方法把 NDCG@10 从 64.0 提升到 70.9，且维度只有原来的八分之一。配合 GPT-5.6-luna 与 DeepSeek-V4.1-Flash，它在 332 个 ToolQA 任务上分别完成 256 和 260 个，超过 BM25、SkillRet-8B、Tool2Vec、RankGPT 等全部基线；在全部 844 个 ToolQA 任务上比最强基线多完成 24 个。在 AppWorld 和 ALFWorld 上则与最强基线持平。论文还指出技能配对会直接影响成败：把任务指南错配到另一个数据库的同级任务，会在 96 个任务中损失 8 个。
CompSkill: Compiling Execution Experience for Skill Retrieval
LLM agents increasingly rely on skill documents, yet most retrievers rank them by description alone and ignore what earlier executions revealed, even though a relevant skill can hurt execution. We present CompSkill, which indexes skills with learned 512-d keys and compiles matched same-task comparisons into skill offsets and capped key residuals (KeyOpt), so experience changes which document the executor receives, never its text. Under a logistic outcome model with unknown task difficulty, matched comparisons cancel that difficulty and only discordant ones carry information; on support executions, predictions improve with each discordant task. An experience hypergraph keeps each comparison with its executor and document versions, recalls missed skills and grounds a cited review. Compact keys raise SkillRet NDCG@10 from 64.0 to 70.9 at one eighth of the dimension. With one document, CompSkill completes 256 and 260 of 332 ToolQA tasks with GPT-5.6-luna and DeepSeek-V4.1-Flash, respectively, more than every baseline, and stays on par with the strongest baselines on AppWorld and ALFWorld. On 332 newly generated ToolQA tasks, it completes 256 with GPT-5.6-luna, significantly more than BM25, SkillRet-8B, Tool2Vec and RankGPT; over all 844 ToolQA tasks, it completes 24 more than the strongest baseline. Selection matters beyond relevance: pairing the right guide with a same-database sibling instead of another database's guide costs 8 of 96 tasks. Code, models and support data: https://github.com/satoshinji2992/CompSkill