---
title: "CompSkill: Compiling Execution Experience for Skill Retrieval"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.01115v1"
permalink: "https://feed.astralor.com/items/itm_276f394ce993481ebc13d2d9"
published: "2026-10-31T17:00:57.000Z"
score: 57
channel: "tech"
tags: ["LLM","AI Agent","OpenAI","RAG","CompSkill","Compiling Execution Experience","Skill Retrieval LLM","KeyOpt"]
---

# CompSkill: Compiling Execution Experience for Skill Retrieval

## 速览

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, whi…

## 关键点

- 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 recei…
- 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.

## 全文

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

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_276f394ce993481ebc13d2d9