paper-with-me

홈 › Papers

Towards Effective Code-Integrated Reasoning

2025-05-30 · Fei Bai, Yingqian Min, Beichen Zhang, Zhipeng Chen, Wayne Xin Zhao, Lei Fang, Zheng Liu, Zhongyuan Wang, Ji-Rong Wen

In this paper, we investigate code-integrated reasoning, where models generate code when necessary and integrate feedback by executing it through a code interpreter. To acquire this capability, models must learn when and how to use external code tools effectively, which is supported by tool-augmented reinforcement learning (RL) through interactive learning. Despite its benefits, tool-augmented RL can still suffer from potential instability in the learning dynamics. In light of this challenge, we present a systematic approach to improving the training effectiveness and stability of tool-augmented RL for code-integrated reasoning. Specifically, we develop enhanced training strategies that balance exploration and stability, progressively building tool-use capabilities while improving reasoning performance. Through extensive experiments on five mainstream mathematical reasoning benchmarks, our model demonstrates significant performance improvements over multiple competitive baselines. Furthermore, we conduct an in-depth analysis of the mechanism and effect of code-integrated reasoning, revealing several key insights, such as the extension of model's capability boundaries and the simultaneous improvement of reasoning efficiency through code integration. All data and code for reproducing this work are available at: https://github.com/RUCAIBox/CIR.

📄 PDF Abstract BibTeX arXiv:2505.24480

Code (1)

rucaibox/cir 공식 구현 pytorch

Tasks

Mathematical ReasoningReinforcement Learning (RL)

Similar Papers 제목 키워드 기반

Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoning

2025-09-27 · Ningning Xu, Yuxuan Jiang, Shubhashis Roy Dipta, Hengyuan Zhang arxiv

Tool-integrated reasoning (TIR) has become a key approach for improving large reasoning models (LRMs) on complex problems. Prior work has mainly studied when to invoke tools, while overlooking how tools are applied. We i…

THOR: Tool-Integrated Hierarchical Optimization via RL for Mathematical Reasoning

2025-09-17 · Qikai Chang, Zhenrong Zhang, Pengfei Hu, Jun Du 외 arxiv

Large Language Models (LLMs) have made remarkable progress in mathematical reasoning, but still continue to struggle with high-precision tasks like numerical computation and formal symbolic manipulation. Integrating exte…

Mathematical ReasoningCode Generation

ToRL: Scaling Tool-Integrated RL

2025-03-30 · Xuefeng Li, Haoyang Zou, PengFei Liu

We introduce ToRL (Tool-Integrated Reinforcement Learning), a framework for training large language models (LLMs) to autonomously use computational tools via reinforcement learning. Unlike supervised fine-tuning, ToRL al…

Mathreinforcement-learningReinforcement Learning

When Do Symbolic Solvers Enhance Reasoning in Large Language Models?

2025-12-02 · Zhiyuan He, Dingmin Wang arxiv

Large Reasoning Models (LRMs) achieve strong performance on complex reasoning tasks by generating long Chains of Thought (CoTs). However, this paradigm might incur substantial token overhead, especially when models "over…

Code Generation

JT-DA: Enhancing Data Analysis with Tool-Integrated Table Reasoning Large Language Models

2025-12-07 · Ce Chi, Xing Wang, Zhendong Wang, Xiaofan Liu 외 arxiv

In this work, we present JT-DA-8B (JiuTian Data Analyst 8B), a specialized large language model designed for complex table reasoning tasks across diverse real-world scenarios. To address the lack of high-quality supervis…

Reinforcement LearningPrompt Engineering