大模型推理能力

Probability-Consistent Preference Optimization for Enhanced LLM Reasoning

Yunqiao Yang, Houxing Ren, Zimu Lu, Ke Wang, Weikang Shi, Aojun Zhou, Junting Pan, Mingjie Zhan, Hongsheng Li

机构 * CUHK MMLab（香港中文大学MML实验室） ； SenseTime Research（商汤科技研究院） ； CPII under InnoHK（创新香港下的CPII） ； Shanghai AI Laboratory（上海人工智能实验室） ； Zhiyuan College, SJTU（上海交通大学智远学院） 纠错

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

Comments 14 pages, to be published in ACL 2025 findings

Patterns Over Principles: The Fragility of Inductive Reasoning in LLMs under Noisy Observations

Chunyang Li, Weiqi Wang, Tianshi Zheng, Yangqiu Song

机构 * Department of Computer Science and Engineering, HKUST, Hong Kong SAR, China（计算机科学与工程系，香港科技大学，香港特别行政区，中国） 纠错

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.AI

Which Data Attributes Stimulate Math and Code Reasoning? An Investigation via Influence Functions

Siqi Kou, Qingyuan Tian, Hanwen Xu, Zihao Zeng, Zhijie Deng

机构 * Shanghai Jiao Tong University（上海交通大学） 纠错

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.LG

SituatedThinker: Grounding LLM Reasoning with Real-World through Situated Thinking

Junnan Liu, Linhao Luo, Thuy-Trang Vu, Gholamreza Haffari

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

Comments Preprint

MMATH: A Multilingual Benchmark for Mathematical Reasoning

Wenyang Luo, Wayne Xin Zhao, Jing Sha, Shijin Wang, Ji-Rong Wen

机构 * Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学首都人工智能学院） ； iFLYTEK Research (Central China), iFLYTEK Co., Ltd.（iFLYTEK中央研究院、iFLYTEK有限公司） 纠错

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

AM-Thinking-v1: Advancing the Frontier of Reasoning at 32B Scale

Yunjie Ji, Xiaoyu Tian, Sitong Zhao, Haotian Wang, Shuaiting Chen, Yiping Peng, Han Zhao, Xiangang Li

机构 * Beike (Ke.com)（贝克（ke.com）） 纠错

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

CER: Confidence Enhanced Reasoning in LLMs

Ali Razghandi, Seyed Mohammad Hadi Hosseini, Mahdieh Soleymani Baghshah

机构 * Sharif University of Technology（谢里夫技术大学） 纠错

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.LG

Comments Accepted at ACL 2025 Main Conference

Thinking Fast and Right: Balancing Accuracy and Reasoning Length with Adaptive Rewards

Jinyan Su, Claire Cardie

机构 * Cornell University（康奈尔大学） 纠错

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

Language Matters: How Do Multilingual Input and Reasoning Paths Affect Large Reasoning Models?

Zhi Rui Tam, Cheng-Kuang Wu, Yu Ying Chiu, Chieh-Yen Lin, Yun-Nung Chen, Hung-yi Lee

机构 * Appier AI Research（Appier人工智能研究院） ； University of Washington（华盛顿大学） ； National Taiwan University（台湾国立大学） 纠错

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

Towards Spoken Mathematical Reasoning: Benchmarking Speech-based Models over Multi-faceted Math Problems

Chengwei Wei, Bin Wang, Jung-jae Kim, Nancy F. Chen

机构 * Institute for Infocomm Research (I 2 R), A*STAR, Singapore（信息与通信研究 institute（I2R），A*STAR，新加坡） ； Centre for Frontier AI Research (CFAR), A*STAR, Singapore（前沿人工智能研究 centre（CFAR），A*STAR，新加坡） 纠错

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

A Survey of Mathematical Reasoning in the Era of Multimodal Large Language Model: Benchmark, Method & Challenges

Yibo Yan, Jiamin Su, Jianxiang He, Fangteng Fu, Xu Zheng, Yuanhuiyi Lyu, Kun Wang, Shen Wang, Qingsong Wen, Xuming Hu

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

Comments Accepted by The 63rd Annual Meeting of the Association for Computational Linguistics (ACL Findings 2025)

Incentivizing Multimodal Reasoning in Large Models for Direct Robot Manipulation

Weiliang Tang, Dong Jing, Jia-Hui Pan, Zhiwu Lu, Yun-Hui Liu, Li Erran Li, Mingyu Ding, Chi-Wing Fu

机构 * Department of Computer Science（计算机科学系） ； The Chinese University of Hong Kong（香港中文大学） ； Gaoling School of Artificial Intelligence（九龙人工智能学院） ； Renmin University of China（中国人民大学） ； AWS AI Amazon（AWS人工智能亚马逊） ； University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校） 纠错

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.AI

Comments 17 pages, 16 figures

Evaluating Mathematical Reasoning Across Large Language Models: A Fine-Grained Approach

Afrar Jahin, Arif Hassan Zidan, Wei Zhang, Yu Bao, Tianming Liu

机构 * School of Computer and Cyber Sciences（计算机与网络安全学院） ； Department of Graduate Psychology（研究生心理学系） ； School of Computing（计算学院） 纠错

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.LG

Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization

Qingyang Zhang, Haitao Wu, Changqing Zhang, Peilin Zhao, Yatao Bian

机构 * Tianjin University（天津大学） ； Tencent AI Lab（腾讯AI实验室） ； NUS（新加坡国立大学） 纠错

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.LG

Comments Ongoing work. First released on April 8, 2025

Token-Efficient Prompt Injection Attack: Provoking Cessation in LLM Reasoning via Adaptive Token Compression

Yu Cui, Yujun Cai, Yiwei Wang

机构 * University of California, Merced（加州大学梅尔德分校） ； University of Queensland（昆士兰大学） 纠错

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.AI

Think, Prune, Train, Improve: Scaling Reasoning without Scaling Models

Caia Costello, Simon Guo, Anna Goldie, Azalia Mirhoseini

机构 * Department of Computer Science, Stanford University（计算机科学系，斯坦福大学） 纠错

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.LG

THOUGHTTERMINATOR: Benchmarking, Calibrating, and Mitigating Overthinking in Reasoning Models

Xiao Pu, Michael Saxon, Wenyue Hua, William Yang Wang

机构 * University of California, Santa Barbara（加州大学圣巴巴拉分校） 纠错

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

CLEAR: Contrasting Textual Feedback with Experts and Amateurs for Reasoning

Andrew Rufail, Daniel Kim, Sean O'Brien, Kevin Zhu

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

Comments Accepted at the Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Student Research Workshop (SRW)

ThoughtProbe: Classifier-Guided Thought Space Exploration Leveraging LLM Intrinsic Reasoning

Zijian Wang, Chang Xu

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

OpenThaiGPT 1.6 and R1: Thai-Centric Open Source and Reasoning Large Language Models

Sumeth Yuenyong, Thodsaporn Chay-intr, Kobkrit Viriyayudhakorn

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

How Difficulty-Aware Staged Reinforcement Learning Enhances LLMs' Reasoning Capabilities: A Preliminary Experimental Study

Yunjie Ji, Sitong Zhao, Xiaoyu Tian, Haotian Wang, Shuaiting Chen, Yiping Peng, Han Zhao, Xiangang Li

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

Large Language Models in Numberland: A Quick Test of Their Numerical Reasoning Abilities

Roussel Rahman

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.AI

Think Twice: Enhancing LLM Reasoning by Scaling Multi-round Test-time Thinking

Xiaoyu Tian, Sitong Zhao, Haotian Wang, Shuaiting Chen, Yunjie Ji, Yiping Peng, Han Zhao, Xiangang Li

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

Long Is More Important Than Difficult for Training Reasoning Models

Si Shen, Fei Huang, Zhixiao Zhao, Chang Liu, Tiansheng Zheng, Danhao Zhu

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

Comments 15 pages,6 figures

LLM Reasoning Engine: Specialized Training for Enhanced Mathematical Reasoning

Shuguang Chen, Guang Lin

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

Comments Accepted to NAACL 2025 KnowledgeNLP

Pensez: Less Data, Better Reasoning -- Rethinking French LLM

Huy Hoang Ha

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

DIMSUM: Discourse in Mathematical Reasoning as a Supervision Module

Krish Sharma, Niyar R Barman, Akshay Chaturvedi, Nicholas Asher

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

Benchmarking Reasoning Robustness in Large Language Models

Tong Yu, Yongcheng Jing, Xikun Zhang, Wentao Jiang, Wenjie Wu, Yingjie Wang, Wenbin Hu, Bo Du, Dacheng Tao

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.AI

Multidimensional Consistency Improves Reasoning in Language Models

Huiyuan Lai, Xiao Zhang, Malvina Nissim

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

Graph-Augmented Reasoning: Evolving Step-by-Step Knowledge Graph Retrieval for LLM Reasoning

Wenjie Wu, Yongcheng Jing, Yingjie Wang, Wenbin Hu, Dacheng Tao

专题命中 数学推理 ：reasoning(title,abstract)；分类 cs.AI