大模型推理能力

From Memorization to Reasoning in the Spectrum of Loss Curvature

Jack Merullo, Srihita Vatsavaya, Lucius Bushnaq, Owen Lewis

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

Towards Understanding Self-play for LLM Reasoning

Justin Yang Chae, Md Tanvirul Alam, Nidhi Rastogi

机构 * University of Washington（华盛顿大学） ； Rochester Institute of Technology（罗切斯特理工学院） 纠错

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

Unlocking Reasoning Capabilities in LLMs via Reinforcement Learning Exploration

Wenhao Deng, Long Wei, Chenglei Yu, Tailin Wu

机构 * Westlake University（西湖大学） 纠错

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

I-RAVEN-X: Benchmarking Generalization and Robustness of Analogical and Mathematical Reasoning in Large Language and Reasoning Models

Giacomo Camposampiero, Michael Hersche, Roger Wattenhofer, Abu Sebastian, Abbas Rahimi

机构 * IBM Research – Zurich（IBM瑞士研究中心） ； ETH Zurich（苏黎世联邦理工学院） 纠错

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

Comments Accepted at the 5th Workshop on Mathematical Reasoning and AI (MATH-AI), NeurIPS 2025

Reinforcing the Diffusion Chain of Lateral Thought with Diffusion Language Models

Zemin Huang, Zhiyang Chen, Zijun Wang, Tiancheng Li, Guo-Jun Qi

机构 * Zhejiang Univeristy（浙江大学） ； MAPLE Lab, Westlake University（西湖大学MAPLE实验室） ； Matterwave Intelligence ； Institute of Advanced Technology, Westlake Institute for Advanced Study（西湖研究所以高级技术研究所） 纠错

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

Comments Accepted to NeurIPS 2025. Code link: https://github.com/maple-research-lab/LLaDOU

ThoughtProbe: Classifier-Guided LLM Thought Space Exploration via Probing Representations

Zijian Wang, Chang Xu

机构 * School of Computer Science（计算机科学学院） ； The University of Sydney（悉尼大学） 纠错

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

Comments EMNLP2025 main conference

Mathematics Isn't Culture-Free: Probing Cultural Gaps via Entity and Scenario Perturbations

Aditya Tomar, Nihar Ranjan Sahoo, Ashish Mittal, Rudra Murthy, Pushpak Bhattacharyya

机构 * IIT Bombay（印度班加罗尔理工学院） ； IBM Research（IBM研究院） 纠错

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

GeoFM: Enhancing Geometric Reasoning of MLLMs via Synthetic Data Generation through Formal Language

Yuhao Zhang, Dingxin Hu, Tinghao Yu, Hao Liu, Yiting Liu

机构 * Tencent Hunyuan Team（腾讯文言团队） 纠错

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

Towards the Formalization of a Trustworthy AI for Mining Interpretable Models explOiting Sophisticated Algorithms

Riccardo Guidotti, Martina Cinquini, Marta Marchiori Manerba, Mattia Setzu, Francesco Spinnato

机构 * University of Pisa（比萨大学） 纠错

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

Building Trustworthy AI by Addressing its 16+2 Desiderata with Goal-Directed Commonsense Reasoning

Alexis R. Tudor, Yankai Zeng, Huaduo Wang, Joaquin Arias, Gopal Gupta

机构 * University of Texas at Dallas（德克萨斯大学达拉斯分校） 纠错

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

HiRA: A Hierarchical Reasoning Framework for Decoupled Planning and Execution in Deep Search

Jiajie Jin, Xiaoxi Li, Guanting Dong, Yuyao Zhang, Yutao Zhu, Yang Zhao, Hongjin Qian, Zhicheng Dou

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

Comments 9 pages

SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agents

Sheng Yin, Xianghe Pang, Yuanzhuo Ding, Menglan Chen, Yutong Bi, Yichen Xiong, Wenhao Huang, Zhen Xiang, Jing Shao, Siheng Chen

机构 * Department of Computer Science and Engineering（计算机科学与工程系） ； Shanghai Jiao Tong University（上海交通大学） ； Department of Computer Science（计算机科学系） ； University of Georgia（佐治亚大学） ； Shanghai AI Laboratory（上海人工智能实验室） 纠错

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

Comments 28 pages, 19 tables, 15 figures

CogPlanner: Unveiling the Potential of Agentic Multimodal Retrieval Augmented Generation with Planning

Xiaohan Yu, Zhihan Yang, Chong Chen

专题命中 规划推理 ：planning(title,abstract)

Comments Accepted by SIGIR-AP 2025

MedCalc-Eval and MedCalc-Env: Advancing Medical Calculation Capabilities of Large Language Models

Kangkun Mao, Jinru Ding, Jiayuan Chen, Mouxiao Bian, Ruiyao Chen, Xinwei Peng, Sijie Ren, Linyang Li, Jie Xu

机构 * Shanghai AI Laboratory（上海人工智能实验室） 纠错

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

RepV: Safety-Separable Latent Spaces for Scalable Neurosymbolic Plan Verification

Yunhao Yang, Neel P. Bhatt, Pranay Samineni, Rohan Siva, Zhanyang Wang, Ufuk Topcu

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

Comments Code and data are available at: https://repv-project.github.io/

ToolScope: An Agentic Framework for Vision-Guided and Long-Horizon Tool Use

Mengjie Deng, Guanting Dong, Zhicheng Dou

机构 * Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学人工智能学院） 纠错

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

KAT-Coder Technical Report

Zizheng Zhan, Ken Deng, Jinghui Wang, Xiaojiang Zhang, Huaixi Tang, Minglei Zhang, Zhiyi Lai, Haoyang Huang, Wen Xiang, Kun Wu, Wenhao Zhuang, Shaojie Wang, Shangpeng Yan, Kepeng Lei, Zongxian Feng, Huiming Wang, Zheng Lin, Mengtong Li, Mengfei Xie, Yinghan Cui, Xuxing Chen, Chao Wang, Weihao Li, Wenqiang Zhu, Jiarong Zhang, Jingxuan Xu, Songwei Yu, Yifan Yao, Xinping Lei, C. Zhang, Han Li, Junqi Xiong, Zuchen Gao, Dailin Li, Haimo Li, Jiaheng Liu, Yuqun Zhang, Junyi Peng, Haotian Zhang, Bin Chen

机构 * Kwaipilot Team（Kwaipilot 团队） 纠错

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

Deep Learning-based Prediction of Clinical Trial Enrollment with Uncertainty Estimates

Tien Huu Do, Antoine Masquelier, Nae Eoun Lee, Jonathan Crowther

机构 * Pfizer（辉瑞公司） ； Merck（默克公司） 纠错

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

MARAG-R1: Beyond Single Retriever via Reinforcement-Learned Multi-Tool Agentic Retrieval

Qi Luo, Xiaonan Li, Yuxin Wang, Tingshuo Fan, Yuan Li, Xinchi Chen, Xipeng Qiu

机构 * School of Computer Science, Fudan University（复旦大学计算机学院） 纠错

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

PoseDiff: A Unified Diffusion Model Bridging Robot Pose Estimation and Video-to-Action Control

Haozhuo Zhang, Michele Caprio, Jing Shao, Qiang Zhang, Jian Tang, Shanghang Zhang, Wei Pan

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

Comments The experimental setup and metrics lacks rigor, affecting the fairness of the comparisons

A Practical-Driven Framework for Transitioning Drive-by-Wire to Autonomous Driving Systems: A Case Study with a Chrysler Pacifica Hybrid Vehicle

Dada Zhang, Md Ruman Islam, Pei-Chi Huang, Chun-Hsing Ho

机构 * Durham School of Architectural Engineering and Construction University of Nebraska-Lincoln（达勒姆建筑工程与建设学院 俄克拉荷马州立大学） ； Department of Computer Science University of Nebraska-Omaha（计算机科学系 俄克拉荷马州立大学） 纠错

专题命中 规划推理 ：planning(abstract)

Comments This updated version includes further implementation details and experimental validation. Accepted for presentation at The 22nd International Conference on Automation Technology (AUTOMATION 2025), Taipei, Taiwan, November 2025

GUI-Rise: Structured Reasoning and History Summarization for GUI Navigation

Tao Liu, Chongyu Wang, Rongjie Li, Yingchen Yu, Xuming He, Bai Song

机构 * ShanghaiTech University（上海科技大学） ； ByteDance（字节跳动） ； Shanghai Engineering Research Center of Intelligent Vision and Imaging（上海智能视觉与成像工程研究中心） 纠错

专题命中 视觉空间推理 ：reasoning(title,abstract)；chain-of-thought(abstract)；分类 cs.AI

Comments Published in NeurIPS 2025

NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation

Xiangyan Liu, Jinjie Ni, Zijian Wu, Chao Du, Longxu Dou, Haonan Wang, Tianyu Pang, Michael Qizhe Shieh

机构 * National University of Singapore（新加坡国立大学） ； Sea AI Lab（Sea AI实验室） 纠错

专题命中 视觉空间推理 ：reasoning(title,abstract)；test-time compute(abstract)

Comments NeurIPS 2025

A Multi-Modal Neuro-Symbolic Approach for Spatial Reasoning-Based Visual Grounding in Robotics

Simindokht Jahangard, Mehrzad Mohammadi, Abhinav Dhall, Hamid Rezatofighi

机构 * Faculty of Information Technology, Monash University（信息技术学院，墨尔本大学） ； Department of Electrical Engineering, Sharif University of Technology（电气工程系，谢赫大学） 纠错

专题命中 视觉空间推理 ：reasoning(title,abstract)；分类 cs.AI

Toward Accurate Long-Horizon Robotic Manipulation: Language-to-Action with Foundation Models via Scene Graphs

Sushil Samuel Dinesh, Shinkyu Park

机构 * Department of Electrical and Computer Engineering, King Abdullah University of Science and Technology (KAUST)（电气与计算机工程系，国王 Abdullah 科学与技术大学） 纠错

专题命中 视觉空间推理 ：reasoning(abstract)；分类 cs.AI、cs.LG

The cell as a token: high-dimensional geometry in language models and cell embeddings

William Gilpin

机构 * Department of Physics, The University of Texas at Austin, Austin, Texas 78712, USA（德克萨斯大学奥斯汀分校物理系） 纠错

专题命中 视觉空间推理 ：reasoning(abstract)；分类 cs.LG

Comments 4 pages, 2 figures

Journal ref Bioinformatics (2025)

VRoPE: Rotary Position Embedding for Video Large Language Models

Zikang Liu, Longteng Guo, Yepeng Tang, Tongtian Yue, Junxian Cai, Kai Ma, Qingbin Liu, Xi Chen, Jing Liu

机构 * Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所） ； School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院） ； School of Computer Science and Technology, Beijing Jiaotong University（北京交通大学计算机科学与技术学院） ； Basic Algorithm Center, Tencent（腾讯基础算法中心） 纠错

专题命中 视觉空间推理 ：reasoning(abstract)；分类 cs.AI

Comments EMNLP 2025 Main Camera Ready

Accelerating Diffusion LLMs via Adaptive Parallel Decoding

Daniel Israel, Guy Van den Broeck, Aditya Grover

专题命中 测试时计算 ：verifier(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments 10 pages, 5 figures

LLMs are Overconfident: Evaluating Confidence Interval Calibration with FermiEval

Elliot L. Epstein, John Winnicki, Thanawat Sornwanee, Rajat Dwaraknath

专题命中 测试时计算 ：reasoning(abstract)；分类 cs.AI、cs.LG

Comments 8 pages

Don't throw the baby out with the bathwater: How and why deep learning for ARC

Jack Cole, Mohamed Osman

机构 * Mindware Consulting, Inc.（Mindware咨询公司） ； MindsAI ； Tufa Labs（Tufa实验室） 纠错

专题命中 测试时计算 ：reasoning(abstract)；分类 cs.AI、cs.LG

Comments 13 pages, 6 figures