大模型推理能力 - arXivDaily 专题

2509.02479 2025-09-04 cs.LG 85%

SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning

Zhenghai Xue, Longtao Zheng, Qian Liu, Yingru Li, Xiaosen Zheng, Zejun Ma, Bo An

机构 * Nanyang Technological University（南洋理工大学）； TikTok

专题命中数学推理：reasoning(title,abstract)；math reasoning(abstract)；self-correction(abstract)；分类 cs.LG

2508.20722 2025-08-29 cs.CL 85%

rStar2-Agent: Agentic Reasoning Technical Report

Ning Shang, Yifei Liu, Yi Zhu, Li Lyna Zhang, Weijiang Xu, Xinyu Guan, Buze Zhang, Bingcheng Dong, Xudong Zhou, Bowen Zhang, Ying Xin, Ziming Miao, Scarlett Li, Fan Yang, Mao Yang

机构 * Microsoft Research（微软研究院）

专题命中数学推理：reasoning(title,abstract)；CoT(abstract)；math reasoning(abstract)；分类 cs.CL

2508.13365 2025-08-20 cs.CL 85%

Stands to Reason: Investigating the Effect of Reasoning on Idiomaticity Detection

Dylan Phelps, Rodrigo Wilkens, Edward Gow-Smith, Thomas Pickard, Maggie Mi, Aline Villavicencio

机构 * School of Computer Science, University of Sheffield, UK（谢菲尔德大学计算机科学学院）； Healthy Lifespan Institute, University of Sheffield（谢菲尔德大学健康寿命研究所）； Institute of Data Science and Artificial Intelligence, University of Exeter, UK（埃克塞特大学数据科学与人工智能研究所）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

2504.00424 2025-08-20 cs.AI 85%

Hawkeye:Efficient Reasoning with Model Collaboration

Jianshu She, Zhuohao Li, Zhemin Huang, Qi Li, Peiran Xu, Haonan Li, Qirong Ho

机构 * Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)（莫扎伊德·本·泽亚德人工智能大学）； University of California, Los Angeles（加州大学洛杉矶分校）； Stanford University（斯坦福大学）； Independent Researcher（独立研究者）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

2507.19748 2025-07-29 cs.CL 85%

JT-Math: A Multi-Stage Framework for Advanced Mathematical Reasoning in Large Language Models

Yifan Hao, Fangning Chao, Yaqian Hao, Zhaojun Cui, Huan Bai, Haiyu Zhang, Yankai Liu, Chao Deng, Junlan Feng

机构 * China Mobile Research Institute Beijing, China（中国移动研究院北京）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

2507.18391 2025-07-25 cs.AI 85%

Revisiting LLM Reasoning via Information Bottleneck

Shiye Lei, Zhihao Cheng, Kai Jia, Dacheng Tao

机构 * The University of Sydney（悉尼大学）； ByteDance（字节跳动）； Nanyang Technological University（南洋理工大学）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

2507.17168 2025-07-24 cs.AI 85%

Improving LLMs' Generalized Reasoning Abilities by Graph Problems

Qifan Zhang, Nuo Chen, Zehua Li, Miao Peng, Jing Tang, Jia Li

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；logical reasoning(abstract)；分类 cs.AI

Comments COLM2025

2507.09850 2025-07-17 cs.AI 85%

The Challenge of Teaching Reasoning to LLMs Without RL or Distillation

Wei Du, Branislav Kisacanin, George Armstrong, Shubham Toshniwal, Ivan Moshkov, Alexan Ayrapetyan, Sadegh Mahdavi, Dan Zhao, Shizhe Diao, Dragan Masulovic, Marius Stanean, Advaith Avadhanam, Max Wang, Ashmit Dutta, Shitij Govil, Sri Yanamandara, Mihir Tandon, Sriram Ananthakrishnan, Vedant Rathi, David Zhang, Joonseok Kang, Leon Luo, Titu Andreescu, Boris Ginsburg, Igor Gitman

机构 * Nvidia Corporation（Nvidia公司）； Institute for Artificial Intelligence Research（人工智能研究机构）； Faculty of Technical Sciences, University of Novi Sad（诺维萨德大学技术科学学院）； Department of Mathematics and Informatics, Faculty of Sciences, University of Novi Sad（诺维萨德大学科学学院数学与信息学系）； University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）； University of Chicago（芝加哥大学）； Georgia Institute of Technology（佐治亚理工学院）； Department of Computer and Information Science, School of Engineering and Applied Science, University of Pennsylvania（宾夕法尼亚大学计算机与信息科学系）； Massachusetts Institute of Technology（麻省理工学院）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

Comments Accepted at the Second AI for Math Workshop at the 42nd International Conference on Machine Learning (ICML 2025)

2507.04391 2025-07-08 cs.CL 85%

Does Learning Mathematical Problem-Solving Generalize to Broader Reasoning?

Ruochen Zhou, Minrui Xu, Shiqi Chen, Junteng Liu, Yunqi Li, Xinxin Lin, Zhengyu Chen, Junxian He

机构 * City University of Hong Kong（香港城市大学）； HKUST（香港科技大学）； Meituan（美团）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

2507.01887 2025-07-03 cs.CL 85%

MiCoTA: Bridging the Learnability Gap with Intermediate CoT and Teacher Assistants

Dongyi Ding, Tiannan Wang, Chenghao Zhu, Meiling Tao, Yuchen Eleanor Jiang, Wangchunshu Zhou

机构 * South China Agricultural University（华南农业大学）； The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））； University of Electronic Science and Technology of China（电子科技大学）

专题命中数学推理：CoT(title,abstract)；reasoning(abstract)；planning(abstract)；分类 cs.CL

Comments Work in progress

2506.12860 2025-06-17 cs.CL 85%

QFFT, Question-Free Fine-Tuning for Adaptive Reasoning

Wanlong Liu, Junxiao Xu, Fei Yu, Yukang Lin, Ke Ji, Wenyu Chen, Yan Xu, Yasheng Wang, Lifeng Shang, Benyou Wang

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments 23 pages

2505.13031 2025-06-12 cs.AI 85%

MindOmni: Unleashing Reasoning Generation in Vision Language Models with RGPO

Yicheng Xiao, Lin Song, Yukang Chen, Yingmin Luo, Yuxin Chen, Yukang Gan, Wei Huang, Xiu Li, Xiaojuan Qi, Ying Shan

机构 * Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院，清华大学）； ARC Lab, Tencent PCG（腾讯PCG ARC实验室）； The Chinese University of Hong Kong（香港中文大学）； The University of Hong Kong（香港大学）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

Comments Code: https://github.com/TencentARC/MindOmni

2505.14652 2025-06-10 cs.CL 85%

General-Reasoner: Advancing LLM Reasoning Across All Domains

Xueguang Ma, Qian Liu, Dongfu Jiang, Ge Zhang, Zejun Ma, Wenhu Chen

机构 * University of Waterloo（多伦多大学）； Vector Institute（向量研究所）； TikTok（TikTok公司）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；verifier(abstract)；分类 cs.CL

2505.23480 2025-05-30 cs.CL 85%

Revisiting Overthinking in Long Chain-of-Thought from the Perspective of Self-Doubt

Keqin Peng, Liang Ding, Yuanxin Ouyang, Meng Fang, Dacheng Tao

机构 * Beihang University（北航大学）； The University of Sydney（悉尼大学）； University of Liverpool（利物浦大学）； Nanyang Technological University（南洋理工大学）

专题命中数学推理：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；分类 cs.CL

2505.10554 2025-05-28 cs.CL 85%

Beyond 'Aha!': Toward Systematic Meta-Abilities Alignment in Large Reasoning Models

Zhiyuan Hu, Yibo Wang, Hanze Dong, Yuhui Xu, Amrita Saha, Caiming Xiong, Bryan Hooi, Junnan Li

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；self-correction(abstract)；分类 cs.CL

Comments In Progress

2505.21178 2025-05-28 cs.CL 85%

Walk Before You Run! Concise LLM Reasoning via Reinforcement Learning

Mingyang Song, Mao Zheng

机构 * Tencent Hunyuan（腾讯文渊）

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments Ongoing Work

2505.20101 2025-05-28 cs.CL 85%

Adaptive Deep Reasoning: Triggering Deep Thinking When Needed

Yunhao Wang, Yuhao Zhang, Tinghao Yu, Can Xu, Feng Zhang, Fengzong Lian

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

2505.14391 2025-05-21 cs.AI 85%

Beyond the First Error: Process Reward Models for Reflective Mathematical Reasoning

Zhaohui Yang, Chenghua He, Xiaowen Shi, Linjing Li, Qiyue Yin, Shihong Deng, Daxin Jiang

机构 * ACL Proceedings（ACL会议）

专题命中数学推理：reasoning(title,abstract)；CoT(abstract)；self-correction(abstract)；分类 cs.AI

2503.19326 2025-04-02 cs.AI 85%

Process or Result? Manipulated Ending Tokens Can Mislead Reasoning LLMs to Ignore the Correct Reasoning Steps

Yu Cui, Bryan Hooi, Yujun Cai, Yiwei Wang

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；self-correction(abstract)；分类 cs.AI

2503.17726 2025-03-25 cs.AI 85%

A Survey on Mathematical Reasoning and Optimization with Large Language Models

Ali Forootani

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；self-correction(abstract)；分类 cs.AI

2503.15268 2025-03-20 cs.AI 85%

Do Chains-of-Thoughts of Large Language Models Suffer from Hallucinations, Cognitive Biases, or Phobias in Bayesian Reasoning?

Roberto Araya

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.AI

Comments 24 pages, 3 figures

2401.17464 2025-01-09 cs.CL 85%

Efficient Tool Use with Chain-of-Abstraction Reasoning

Silin Gao, Jane Dwivedi-Yu, Ping Yu, Xiaoqing Ellen Tan, Ramakanth Pasunuru, Olga Golovneva, Koustuv Sinha, Asli Celikyilmaz, Antoine Bosselut, Tianlu Wang

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；planning(abstract)；分类 cs.CL

2401.08967 2024-12-16 cs.CL 85%

ReFT: Reasoning with Reinforced Fine-Tuning

Trung Quoc Luong, Xinbo Zhang, Zhanming Jie, Peng Sun, Xiaoran Jin, Hang Li

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments ACL 2024 main conference; adjust with reviewer comments; 13 pages

2411.05407 2024-11-11 cs.CL 85%

Gap-Filling Prompting Enhances Code-Assisted Mathematical Reasoning

Mohammad Ghiasvand Mohammadkhani

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

2409.20441 2024-11-01 cs.CL 85%

Instance-adaptive Zero-shot Chain-of-Thought Prompting

Xiaosong Yuan, Chen Shen, Shaotian Yan, Xiaofeng Zhang, Liang Xie, Wenxiao Wang, Renchu Guan, Ying Wang, Jieping Ye

专题命中数学推理：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；分类 cs.CL

Comments Accepted by NeurIPS 2024

2407.03624 2024-08-27 cs.CL 85%

Question-Analysis Prompting Improves LLM Performance in Reasoning Tasks

Dharunish Yugeswardeenoo, Kevin Zhu, Sean O'Brien

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments Accepted in Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics: Student Research Workshop (ACL-SRW 2024) 11 pages, 8 figures

2302.12246 2024-07-23 cs.CL 85%

Active Prompting with Chain-of-Thought for Large Language Models

Shizhe Diao, Pengcheng Wang, Yong Lin, Rui Pan, Xiang Liu, Tong Zhang

专题命中数学推理：chain-of-thought(title,abstract)；reasoning(abstract)；CoT(abstract)；分类 cs.CL

Comments Published in ACL 2024

2404.17140 2024-06-07 cs.CL 85%

Small Language Models Need Strong Verifiers to Self-Correct Reasoning

Yunxiang Zhang, Muhammad Khalifa, Lajanugen Logeswaran, Jaekyeom Kim, Moontae Lee, Honglak Lee, Lu Wang

专题命中数学推理：reasoning(title,abstract)；verifier(abstract)；self-correction(abstract)；分类 cs.CL

Comments ACL Findings 2024 - Camera Ready

2306.06427 2024-06-04 cs.CL 85%

Boosting Language Models Reasoning with Chain-of-Knowledge Prompting

Jianing Wang, Qiushi Sun, Xiang Li, Ming Gao

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments ACL 2024

2308.09267 2024-04-23 cs.AI 85%

GraphReason: Enhancing Reasoning Capabilities of Large Language Models through A Graph-Based Verification Approach

Lang Cao

专题命中数学推理：reasoning(title,abstract)；chain-of-thought(abstract)；verifier(abstract)；分类 cs.AI

1. 数学推理 3208 篇

SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning

rStar2-Agent: Agentic Reasoning Technical Report

Stands to Reason: Investigating the Effect of Reasoning on Idiomaticity Detection

Hawkeye:Efficient Reasoning with Model Collaboration

JT-Math: A Multi-Stage Framework for Advanced Mathematical Reasoning in Large Language Models

Revisiting LLM Reasoning via Information Bottleneck

Improving LLMs' Generalized Reasoning Abilities by Graph Problems

The Challenge of Teaching Reasoning to LLMs Without RL or Distillation

Does Learning Mathematical Problem-Solving Generalize to Broader Reasoning?

MiCoTA: Bridging the Learnability Gap with Intermediate CoT and Teacher Assistants

QFFT, Question-Free Fine-Tuning for Adaptive Reasoning

MindOmni: Unleashing Reasoning Generation in Vision Language Models with RGPO

General-Reasoner: Advancing LLM Reasoning Across All Domains

Revisiting Overthinking in Long Chain-of-Thought from the Perspective of Self-Doubt

Beyond 'Aha!': Toward Systematic Meta-Abilities Alignment in Large Reasoning Models

Walk Before You Run! Concise LLM Reasoning via Reinforcement Learning

Adaptive Deep Reasoning: Triggering Deep Thinking When Needed

Beyond the First Error: Process Reward Models for Reflective Mathematical Reasoning

Process or Result? Manipulated Ending Tokens Can Mislead Reasoning LLMs to Ignore the Correct Reasoning Steps

A Survey on Mathematical Reasoning and Optimization with Large Language Models

Do Chains-of-Thoughts of Large Language Models Suffer from Hallucinations, Cognitive Biases, or Phobias in Bayesian Reasoning?

Efficient Tool Use with Chain-of-Abstraction Reasoning

ReFT: Reasoning with Reinforced Fine-Tuning

Gap-Filling Prompting Enhances Code-Assisted Mathematical Reasoning

Instance-adaptive Zero-shot Chain-of-Thought Prompting

Question-Analysis Prompting Improves LLM Performance in Reasoning Tasks

Active Prompting with Chain-of-Thought for Large Language Models

Small Language Models Need Strong Verifiers to Self-Correct Reasoning

Boosting Language Models Reasoning with Chain-of-Knowledge Prompting

GraphReason: Enhancing Reasoning Capabilities of Large Language Models through A Graph-Based Verification Approach