大模型推理能力

Staying in the Sweet Spot: Responsive Reasoning Evolution via Capability-Adaptive Hint Scaffolding

Ziheng Li, Zexu Sun, Jinman Zhao, Erxue Min, Yongcheng Zeng, Hui Wu, Hengyi Cai, Shuaiqiang Wang, Dawei Yin, Xu Chen, Zhi-Hong Deng

机构 * School of Intelligence Science and Technology, Peking University（北京大学智能科学与技术学院） ； Baidu Inc.（百度公司） ； Department of Computer Science, University of Toronto（多伦多大学计算机科学系） ； Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所） ； Aerospace Information Research Institute, Chinese Academy of Sciences（中国科学院航天信息研究所） ； Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学光荣人工智能学院） 纠错

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

Comments Work in progress

Outcome-based Exploration for LLM Reasoning

Yuda Song, Julia Kempe, Remi Munos

机构 * FAIR at Meta（Meta 的 FAIR） 纠错

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

Comments 26 pages, 11 figures

A Fragile Number Sense: Probing the Elemental Limits of Numerical Reasoning in LLMs

Roussel Rahman, Aashwin Ananda Mishra

机构 * Stanford University（斯坦福大学） ； SLAC National Accelerator Laboratory（SLAC国家加速器实验室） 纠错

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

From Implicit Exploration to Structured Reasoning: Leveraging Guideline and Refinement for LLMs

Jiaxiang Chen, Zhuo Wang, Mingxi Zou, Zhucong Li, Zhijian Zhou, Song Wang, Zenglin Xu

机构 * Fudan University（复旦大学） ； Shanghai Innovation Institute（上海创新研究院） ； Zoom Video Communications（Zoom视频通讯） 纠错

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

Multimodal Reasoning for Science: Technical Report and 1st Place Solution to the ICML 2025 SeePhys Challenge

Hao Liang, Ruitao Wu, Bohan Zeng, Junbo Niu, Wentao Zhang, Bin Dong

机构 * Peking University（北京大学） ； Beihang University（北航） ； Zhongguancun Academy（中关村学院） 纠错

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

Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models

Yinjie Wang, Ling Yang, Bowen Li, Ye Tian, Ke Shen, Mengdi Wang

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

Comments Code and Models: https://github.com/Gen-Verse/dLLM-RL

Large Language Models Might Not Care What You Are Saying: Prompt Format Beats Descriptions

Chenming Tang, Zhixiang Wang, Hao Sun, Yunfang Wu

机构 * National Key Laboratory for Multimedia Information Processing, Peking University（北京大学多媒体信息处理国家重点实验室） ； MOE Key Laboratory of Computational Linguistics, Peking University School of Computer Science, Peking University（教育部计算语言学重点实验室，北京大学计算机科学学院，北京大学） 纠错

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

Comments EMNLP 2025 Findings. 23 pages, 23 figures, 7 tables

DCPO: Dynamic Clipping Policy Optimization

Shihui Yang, Chengfeng Dou, Peidong Guo, Kai Lu, Qiang Ju, Fei Deng, Rihui Xin

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

Learning to Reason for Long-Form Story Generation

Alexander Gurung, Mirella Lapata

机构 * School of Informatics University of Edinburgh（信息学院爱丁堡大学） 纠错

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

Error Classification of Large Language Models on Math Word Problems: A Dynamically Adaptive Framework

Yuhong Sun, Zhangyue Yin, Xuanjing Huang, Xipeng Qiu, Hui Zhao

机构 * Software Engineering Institute, East China Normal University（东华师范大学软件工程研究院） ； School of Computer Science and Artificial Intelligence, Fudan University（复旦大学计算机科学与人工智能学院） ； Shanghai Key Laboratory of Trustworthy Computing, Shanghai, China（上海可信计算关键实验室） 纠错

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

Comments 28 pages, 10 figures, accepted by Findings of EMNLP2025

Saturation-Driven Dataset Generation for LLM Mathematical Reasoning in the TPTP Ecosystem

Valentin Quesnel, Damien Sileo

专题命中 代码与定理证明 ：reasoning(title,abstract)；分类 cs.CL、cs.AI

ReST-RL: Achieving Accurate Code Reasoning of LLMs with Optimized Self-Training and Decoding

Sining Zhoubian, Dan Zhang, Jie Tang

机构 * The Knowledge Engineering Group (KEG), Tsinghua University（清华大学知识工程小组（KEG）、清华大学） 纠错

专题命中 代码与定理证明 ：reasoning(title,abstract)；分类 cs.AI、cs.LG

Comments 21 pages, 4 figures

Proof-Carrying Numbers (PCN): A Protocol for Trustworthy Numeric Answers from LLMs via Claim Verification

Aivin V. Solatorio

专题命中 代码与定理证明 ：verifier(abstract)；分类 cs.CL、cs.LG

From Long to Short: LLMs Excel at Trimming Own Reasoning Chains

Wei Han, Geng Zhan, Sicheng Yu, Chenyu Wang, Bryan Hooi

专题命中 逻辑推理 ：reasoning(title,abstract)；分类 cs.CL、cs.AI

Comments 21 pages, 5 figures, 7 tables

DRF: LLM-AGENT Dynamic Reputation Filtering Framework

Yuwei Lou, Hao Hu, Shaocong Ma, Zongfei Zhang, Liang Wang, Jidong Ge, Xianping Tao

机构 * State Key Laboratory for Novel Software Technology（新型软件技术国家重点实验室） ； Nanjing University（南京大学） ； Amazon.com Services LLC（亚马逊公司） 纠错

专题命中 逻辑推理 ：reasoning(abstract)；logical reasoning(abstract)；分类 cs.AI

Comments This paper has been accepted by ICONIP 2025 but not published

Presolving Convexified Optimal Power Flow with Mixtures of Gradient Experts

Shourya Bose, Kejun Chen, Yu Zhang

专题命中 逻辑推理 ：planning(abstract)

Comments Accepted in "Energy and AI"

Robotic Fire Risk Detection based on Dynamic Knowledge Graph Reasoning: An LLM-Driven Approach with Graph Chain-of-Thought

Haimei Pan, Jiyun Zhang, Qinxi Wei, Xiongnan Jin, Chen Xinkai, Jie Cheng

专题命中 规划推理 ：reasoning(title)；chain-of-thought(title)；planning(abstract)；分类 cs.AI

Comments We have decided to withdraw this paper as the work is still undergoing further refinement. To ensure the clarity of the results, we prefer to make additional improvements before resubmission. We appreciate the readers' understanding

Planning with Reasoning using Vision Language World Model

Delong Chen, Theo Moutakanni, Willy Chung, Yejin Bang, Ziwei Ji, Allen Bolourchi, Pascale Fung

机构 * Meta FAIR 纠错

专题命中 规划推理 ：planning(title,abstract)；reasoning(title)；分类 cs.AI

Scaling Laws of Motion Forecasting and Planning -- Technical Report

Mustafa Baniodeh, Kratarth Goel, Scott Ettinger, Carlos Fuertes, Ari Seff, Tim Shen, Cole Gulino, Chenjie Yang, Ghassen Jerfel, Dokook Choe, Rui Wang, Benjamin Charrow, Vinutha Kallem, Sergio Casas, Rami Al-Rfou, Benjamin Sapp, Dragomir Anguelov

机构 * Waymo LLC（Waymo公司） 纠错

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI、cs.LG

Efficient Virtuoso: A Latent Diffusion Transformer Model for Goal-Conditioned Trajectory Planning

Antonio Guillen-Perez

机构 * Independent Researcher（独立研究者） 纠错

专题命中 规划推理 ：planning(title,abstract)；分类 cs.AI、cs.LG

Deep Reactive Policy: Learning Reactive Manipulator Motion Planning for Dynamic Environments

Jiahui Yang, Jason Jingzhou Liu, Yulong Li, Youssef Khaky, Kenneth Shaw, Deepak Pathak

机构 * Carnegie Mellon University（卡内基梅隆大学） 纠错

专题命中 规划推理 ：planning(title)；分类 cs.AI、cs.LG

Comments Website at \url{deep-reactive-policy.com}

REMI: A Novel Causal Schema Memory Architecture for Personalized Lifestyle Recommendation Agents

Vishal Raman, Vijai Aravindh R, Abhijith Ragav

机构 * Radian Group Inc.（Radian集团） ； Sri Sivasubramaniya Nadar College Of Engineering（Sri Sivasubramaniya纳达尔工程学院） ； Amazon（亚马逊） 纠错

专题命中 规划推理 ：reasoning(abstract)；planning(abstract)；分类 cs.AI

Comments 8 pages, 2 figures, Accepted at the OARS Workshop, KDD 2025, Paper link: https://oars-workshop.github.io/papers/Raman2025.pdf

Affective Computing in the Era of Large Language Models: A Survey from the NLP Perspective

Yiqun Zhang, Xiaocui Yang, Xingle Xu, Zeran Gao, Yijie Huang, Shiyi Mu, Shi Feng, Daling Wang, Yifei Zhang, Kaisong Song, Ge Yu

机构 * Northeastern University, China（东北大学） ； Alibaba Group, Hangzhou, China（阿里巴巴集团） 纠错

专题命中 规划推理 ：chain-of-thought(abstract)；planning(abstract)；分类 cs.CL

Comments Compared with the previous version, reinforcement learning has been added (as a new section), including RLHF, RLVR, and RLAIF

SpecSwin3D: Generating Hyperspectral Imagery from Multispectral Data via Transformer Networks

Tang Sui, Songxi Yang, Qunying Huang

机构 * Department of Geography, University of Wisconsin-Madison（威斯康星大学麦迪逊分校地理系） 纠错

专题命中 规划推理 ：planning(abstract)；分类 cs.AI

MAPF-World: Action World Model for Multi-Agent Path Finding

Zhanjiang Yang, Yang Shen, Yueming Li, Meng Li, Lijun Sun

机构 * Shenzhen Technology University（深圳科技大学） ； University of Technology Sydney（悉尼技术大学） 纠错

专题命中 规划推理 ：planning(abstract)；分类 cs.AI

Integrating Evidence into the Design of XAI and AI-based Decision Support Systems: A Means-End Framework for End-users in Construction

Peter E. D. Love, Jane Matthews, Weili Fang, Hadi Mahamivanan

专题命中 规划推理 ：reasoning(abstract)；分类 cs.AI

Comments 74 pages, 5 figures and 3 tables

GameGPT: Multi-agent Collaborative Framework for Game Development

Dake Chen, Haoyang Zhang, Hanbin Wang, Yunhao Huo, Yuzhao Li, Junjie Wang

机构 * AutoGame Research（AutoGame研究所） ； X-Institute（X研究所） ； University of Southern California（南加州大学） ； Tsinghua University（清华大学） 纠错

专题命中 规划推理 ：planning(abstract)；分类 cs.AI

Reinforcement Learning with Anticipation: A Hierarchical Approach for Long-Horizon Tasks

Yang Yu

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

专题命中 规划推理 ：planning(abstract)；分类 cs.LG

From Staff Messages to Actionable Insights: A Multi-Stage LLM Classification Framework for Healthcare Analytics

Hajar Sakai, Yi-En Tseng, Mohammadsadegh Mikaeili, Joshua Bosire, Franziska Jovin

专题命中 规划推理 ：reasoning(abstract)；分类 cs.CL

LLMs in Cybersecurity: Friend or Foe in the Human Decision Loop?

Irdin Pekaric, Philipp Zech, Tom Mattson

专题命中 规划推理 ：reasoning(abstract)