大模型推理能力

Wait, that's not an option: LLMs Robustness with Incorrect Multiple-Choice Options

Gracjan Góral, Emilia Wiśnios, Piotr Sankowski, Paweł Budzianowski

机构 * University of Warsaw（华沙大学） ； Institute of Mathematics, Polish Academy of Sciences（波兰科学院数学研究所） ； MIM Solutions（MIM解决方案） ； K-Scale Labs（K-Scale实验室） ； IDEAS NCBR ； IDEAS Research Institute（IDEAS研究学院） 纠错

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

Comments Accepted for ACL 2025 Main Conference and NeurIPS 2024 FM-EduAssess Workshop

LEMMA: Learning from Errors for MatheMatical Advancement in LLMs

Zhuoshi Pan, Yu Li, Honglin Lin, Qizhi Pei, Zinan Tang, Wei Wu, Chenlin Ming, H. Vicky Zhao, Conghui He, Lijun Wu

机构 * Tsinghua University（清华大学） ； Shanghai AI Laboratory（上海人工智能实验室） ； Renmin University of China（中国人民大学） ； University of Science and Technology of China（中国科学技术大学） ； Shanghai Jiao Tong University（上海交通大学） 纠错

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

Comments ACL'25 Findings, Code is available at https://github.com/pzs19/LEMMA

Who Reasons in the Large Language Models?

Jie Shao, Jianxin Wu

机构 * National Key Laboratory for Novel Software Technology, Nanjing University, China（国家新型软件技术实验室，南京大学，中国） ； School of Artificial Intelligence, Nanjing University, China（人工智能学院，南京大学，中国） 纠错

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

SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training

Tianzhe Chu, Yuexiang Zhai, Jihan Yang, Shengbang Tong, Saining Xie, Dale Schuurmans, Quoc V. Le, Sergey Levine, Yi Ma

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

Comments Website at https://tianzhechu.com/SFTvsRL

Why Vision Language Models Struggle with Visual Arithmetic? Towards Enhanced Chart and Geometry Understanding

Kung-Hsiang Huang, Can Qin, Haoyi Qiu, Philippe Laban, Shafiq Joty, Caiming Xiong, Chien-Sheng Wu

机构 * Salesforce AI Research（Salesforce AI研究院） ； UCLA（加州大学洛杉矶分校） 纠错

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

Comments Code and data are available at https://github.com/SalesforceAIResearch/CogAlign

Assessing Robustness to Spurious Correlations in Post-Training Language Models

Julia Shuieh, Prasann Singhal, Apaar Shanker, John Heyer, George Pu, Samuel Denton

机构 * Scale AI 纠错

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

Comments ICLR '25 Workshop on Spurious Correlation and Shortcut Learning

CHORUS: Zero-shot Hierarchical Retrieval and Orchestration for Generating Linear Programming Code

Tasnim Ahmed, Salimur Choudhury

机构 * School of Computing, Queen’s University, Ontario, Canada（计算学院，皇后大学，安大略省，加拿大） 纠错

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

Comments This paper has been accepted for presentation at the 19th Learning and Intelligent Optimization Conference (LION 19)

BitNet b1.58 2B4T Technical Report

Shuming Ma, Hongyu Wang, Shaohan Huang, Xingxing Zhang, Ying Hu, Ting Song, Yan Xia, Furu Wei

机构 * Microsoft Research（微软研究院） ； University of Chinese Academy of Sciences（中国科学院大学） ； Tsinghua University（清华大学） 纠错

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

Comments Work in progress

MIND: Math Informed syNthetic Dialogues for Pretraining LLMs

Syeda Nahida Akter, Shrimai Prabhumoye, John Kamalu, Sanjeev Satheesh, Eric Nyberg, Mostofa Patwary, Mohammad Shoeybi, Bryan Catanzaro

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

Comments 31 pages, 5 figures, 14 tables

A Post-trainer's Guide to Multilingual Training Data: Uncovering Cross-lingual Transfer Dynamics

Luisa Shimabucoro, Ahmet Ustun, Marzieh Fadaee, Sebastian Ruder

机构 * Cohere For AI ； University of São Paulo（圣保罗大学） ； Meta 纠错

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

Brains vs. Bytes: Evaluating LLM Proficiency in Olympiad Mathematics

Hamed Mahdavi, Alireza Hashemi, Majid Daliri, Pegah Mohammadipour, Alireza Farhadi, Samira Malek, Yekta Yazdanifard, Amir Khasahmadi, Vasant Honavar

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

From Blind Solvers to Logical Thinkers: Benchmarking LLMs' Logical Integrity on Faulty Mathematical Problems

A M Muntasir Rahman, Junyi Ye, Wei Yao, Sierra S. Liu, Jesse Yu, Jonathan Yu, Wenpeng Yin, Guiling Wang

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

LLM Library Learning Fails: A LEGO-Prover Case Study

Ian Berlot-Attwell, Frank Rudzicz, Xujie Si

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

Comments 24 pages, 5 figures

On Vanishing Variance in Transformer Length Generalization

Ruining Li, Gabrijel Boduljak, Jensen, Zhou

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

Comments Project page: https://ruiningli.com/vanishing-variance. The first two authors contributed equally to this work

MAQA: Evaluating Uncertainty Quantification in LLMs Regarding Data Uncertainty

Yongjin Yang, Haneul Yoo, Hwaran Lee

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

Comments Findings of NAACL 2025

Solving Situation Puzzles with Large Language Model and External Reformulation

Kun Li, Xinwei Chen, Tianyou Song, Chengrui Zhou, Zhuoran Liu, Zhenyan Zhang, Jiangjian Guo, Qing Shan

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

Token-Supervised Value Models for Enhancing Mathematical Problem-Solving Capabilities of Large Language Models

Jung Hyun Lee, June Yong Yang, Byeongho Heo, Dongyoon Han, Kyungsu Kim, Eunho Yang, Kang Min Yoo

专题命中 数学推理 ：test-time compute(abstract)；分类 cs.CL、cs.AI

Process-based Self-Rewarding Language Models

Shimao Zhang, Xiao Liu, Xin Zhang, Junxiao Liu, Zheheng Luo, Shujian Huang, Yeyun Gong

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

Universal AI maximizes Variational Empowerment

Yusuke Hayashi, Koichi Takahashi

专题命中 数学推理 ：planning(abstract)；分类 cs.AI、cs.LG

Comments 9 pages, no figures

When an LLM is apprehensive about its answers -- and when its uncertainty is justified

Petr Sychev, Andrey Goncharov, Daniil Vyazhev, Edvard Khalafyan, Alexey Zaytsev

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

Language Imbalance Driven Rewarding for Multilingual Self-improving

Wen Yang, Junhong Wu, Chen Wang, Chengqing Zong, Jiajun Zhang

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

Comments Camera ready version for ICLR 2025

Concept Distillation from Strong to Weak Models via Hypotheses-to-Theories Prompting

Emmanuel Aboah Boateng, Cassiano O. Becker, Nabiha Asghar, Kabir Walia, Ashwin Srinivasan, Ehi Nosakhare, Soundar Srinivasan, Victor Dibia

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

Comments Accepted to NAACL 2025; 17 pages, 8 figures

A novel approach to the relationships between data features -- based on comprehensive examination of mathematical, technological, and causal methodology

JaeHong Kim

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

Comments 59 pages, 6 figures, 2 tables

Is Free Self-Alignment Possible?

Dyah Adila, Changho Shin, Yijing Zhang, Frederic Sala

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

Sens-Merging: Sensitivity-Guided Parameter Balancing for Merging Large Language Models

Shuqi Liu, Han Wu, Bowei He, Xiongwei Han, Mingxuan Yuan, Linqi Song

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

MergeME: Model Merging Techniques for Homogeneous and Heterogeneous MoEs

Yuhang Zhou, Giannis Karamanolakis, Victor Soto, Anna Rumshisky, Mayank Kulkarni, Furong Huang, Wei Ai, Jianhua Lu

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

Comments Accepted by NAACL 2025 Main

LLMs for Mathematical Modeling: Towards Bridging the Gap between Natural and Mathematical Languages

Xuhan Huang, Qingning Shen, Yan Hu, Anningzhe Gao, Benyou Wang

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

Comments Findings of NAACL2025. Project: https://github.com/FreedomIntelligence/Mamo

OpenGrok: Enhancing SNS Data Processing with Distilled Knowledge and Mask-like Mechanisms

Lumen AI, Zaozhuang No. 28 Middle School, Shihao Ji, Zihui Song, Fucheng Zhong, Jisen Jia, Zhaobo Wu, Zheyi Cao, Tianhao Xu

专题命中 数学推理 ：CoT(abstract)；分类 cs.AI、cs.LG

Comments 7 pages

Causal Inference with Large Language Model: A Survey

Jing Ma

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

Comments 12 pages, 2 figures, 4 tables

LLM The Genius Paradox: A Linguistic and Math Expert's Struggle with Simple Word-based Counting Problems

Nan Xu, Xuezhe Ma

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

Comments NAACL 2025