---
title: "Think with Skills: Pseudocode Planning Before Execution for Advanced Reasoners"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.02830v1"
permalink: "https://feed.astralor.com/items/itm_4c6412fc2f6804ecde35239f"
published: "2026-11-02T09:29:30.000Z"
score: 57
channel: "ai"
tags: ["AI Agent","Think","Skills","Pseudocode Planning Before","Execution","Advanced Reasoners Language","Such","Language"]
---

# Think with Skills: Pseudocode Planning Before Execution for Advanced Reasoners

## 速览

Language model agents often struggle with complex tasks even when equipped with expert-written skills. Such skills encode declarative knowledge, whereas successful execution requires procedural knowledge tailored to the…

## 关键点

- Language model agents often struggle with complex tasks even when equipped with expert-written skills.
- Such skills encode declarative knowledge, whereas successful execution requires procedural knowledge tailored to the task at hand.
- Natural language skills may leave conditions and dependencies implicit, and tasks involving multiple skills further require agents to coordinate intermediate results and execution order.

## 全文

Language model agents often struggle with complex tasks even when equipped with expert-written skills. Such skills encode declarative knowledge, whereas successful execution requires procedural knowledge tailored to the task at hand. Natural language skills may leave conditions and dependencies implicit, and tasks involving multiple skills further require agents to coordinate intermediate results and execution order. We introduce Think with Skills, a method that trains reasoning models to turn relevant skill descriptions into task-specific pseudocode plans before execution. The pseudocode explicitly represents operations, conditions, intermediate results, and dependencies, providing a structured bridge from skill descriptions to agent actions. We construct training data that pairs tasks and skills with pseudocode plans and execution trajectories, use supervised fine tuning to initialize this behavior, and further optimize end-to-end task performance with Group Relative Policy Optimization. On SkillsBench, Think with Skills raises task success from 35.0% to 62.0%. These results show that proceduralizing expert knowledge before execution substantially improves skill use.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_4c6412fc2f6804ecde35239f