大模型推理能力

The Impact of Quantization on Large Reasoning Model Reinforcement Learning

量化对大推理模型强化学习的影响

Medha Kumar, Zifei Xu, Xin Wang, Tristan Webb

机构 * Pennsylvania State University（宾夕法尼亚州立大学） ； d-Matrix 纠错

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

AI总结 本文研究了量化对大推理模型强化学习的影响，发现量化感知训练对学习过程有负面影响，而PTQ和QLoRA能提升性能。

Comments Accepted to the NeurIPS 2025 Efficient Reasoning Workshop

DEPO: Dual-Efficiency Preference Optimization for LLM Agents

DEPO：用于LLM代理的双效偏好优化

Sirui Chen, Mengshi Zhao, Lei Xu, Yuying Zhao, Beier Zhu, Hanwang Zhang, Shengjie Zhao, Chaochao Lu

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

AI总结 DEPO通过双效偏好优化方法提升LLM代理的效率和性能，减少token和步骤消耗，同时在多个基准测试中表现优异。

Comments Accepted to AAAI 2026

Two-Faced Social Agents: Context Collapse in Role-Conditioned Large Language Models

Vikram K Suresh

专题命中 数学推理 ：reasoning(abstract)

Socrates or Smartypants: Testing Logic Reasoning Capabilities of Large Language Models with Logic Programming-based Test Oracles

Zihao Xu, Junchen Ding, Yiling Lou, Kun Zhang, Dong Gong, Yuekang Li

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

ProRAC: A Neuro-symbolic Method for Reasoning about Actions with LLM-based Progression

Haoyong Wu, Yongmei Liu

机构 * Haoyong Wu, Yongmei Liu（作者） 纠错

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

When Words Change the Model: Sensitivity of LLMs for Constraint Programming Modelling

Alessio Pellegrino, Jacopo Mauro

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

Planning-Aware Code Infilling via Horizon-Length Prediction

Yifeng Ding, Hantian Ding, Shiqi Wang, Qing Sun, Varun Kumar, Zijian Wang

机构 * University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校） ； AWS AI Labs（AWS人工智能实验室） ； Meta 纠错

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

RRT*former: Environment-Aware Sampling-Based Motion Planning using Transformer

RRT*former：基于变换器的环境感知运动规划

Mingyang Feng, Shaoyuan Li, Xiang Yin

机构 * School of Automation and Intelligent Sensing, Shanghai Jiao Tong University（自动化与智能感知学院，上海交通大学） 纠错

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

AI总结 RRT*former 是一种结合变换器网络的新型运动规划算法，通过利用环境信息和先前样本提升路径规划的最优性和效率。

Comments Accepted to IROS 2025

Path Planning through Multi-Agent Reinforcement Learning in Dynamic Environments

Jonas De Maeyer, Hossein Yarahmadi, Moharram Challenger

机构 * Department of Computer Science University of Antwerp (UA)（安特卫普大学计算机科学系） ； Department of Computer Engineering, Faculty of Engineering, Ayatollah Boroujerdi University（阿亚图拉·博鲁杰尔迪大学工程学院计算机工程系） ； Department of Computer Science University of Antwerp (UA) and Flanders Make（安特卫普大学计算机科学系和弗拉芒制造） 纠错

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

Driving with Regulation: Trustworthy and Interpretable Decision-Making for Autonomous Driving with Retrieval-Augmented Reasoning

Tianhui Cai, Yifan Liu, Zewei Zhou, Haoxuan Ma, Seth Z. Zhao, Zhiwen Wu, Xu Han, Zhiyu Huang, Jiaqi Ma

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

MedBuild AI: An Agent-Based Hybrid Intelligence Framework for Reshaping Agency in Healthcare Infrastructure Planning through Generative Design for Medical Architecture

Yiming Zhang, Yuejia Xu, Ziyao Wang, Xin Yan, Xiaosai Hao

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

Comments 24 pages, 16 figures. Submitted to the IJAC Special Issue "Rebalance and Reciprocity"

An LLM-Powered Agent for Real-Time Analysis of the Vietnamese IT Job Market

Minh-Thuan Nguyen, Thien Vo-Thanh, Thai-Duy Dinh, Xuan-Quang Phan, Tan-Ha Mai, Lam-Son Lê

机构 * Computer Science Department Vietnamese-German University, Vietnam（越南德意志大学计算机科学系） ； Business Administration Department FPT University, Vietnam（越南FPT大学商学院） ； IT Operation Department Mantu Group, Vietnam（越南Mantu集团IT运营部） ； CSIE Department National Taiwan University, Taiwan（台湾国立台湾大学CSIE系） 纠错

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

Comments Accepted at ACOMPA 2025

PropRAG: Guiding Retrieval with Beam Search over Proposition Paths

Jingjin Wang, Jiawei Han

机构 * Siebel School of Computing and Data Science, University of Illinois Urbana-Champaign（塞比尔计算与数据科学学院，伊利诺伊大学厄巴纳-香槟分校） 纠错

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

Comments Accepted to EMNLP 2025 (Main Conference). Camera-ready version. Code and data: https://github.com/ReLink-Inc/PropRAG

Journal ref Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 6224-6239 November 4-9, 2025

Agent-SAMA: State-Aware Mobile Assistant

Linqiang Guo, Wei Liu, Yi Wen Heng, Tse-Hsun, Chen, Yang Wang

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

Comments Accepted to AAAI-26 (Main Technical Track)

When CNNs Outperform Transformers and Mambas: Revisiting Deep Architectures for Dental Caries Segmentation

Aashish Ghimire, Jun Zeng, Roshan Paudel, Nikhil Kumar Tomar, Deepak Ranjan Nayak, Harshith Reddy Nalla, Vivek Jha, Glenda Reynolds, Debesh Jha

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

Comments 8 pages, 4 figures

Real-Time Optimal Control via Transformer Networks and Bernstein Polynomials

Gage MacLin, Venanzio Cichella, Andrew Patterson, Irene Gregory

专题命中 规划推理 ：planning(abstract)

Comments Accepted to 2025 IEEE Conference on Decision and Control (CDC)

What About Emotions? Guiding Fine-Grained Emotion Extraction from Mobile App Reviews

Quim Motger, Marc Oriol, Max Tiessler, Xavier Franch, Jordi Marco

专题命中 规划推理 ：planning(abstract)

Comments Accepted at the 33rd IEEE International Requirements Engineering 2025 conference

Simulating an Autonomous System in CARLA using ROS 2

Joseph Abdo, Aditya Shibu, Moaiz Saeed, Abdul Maajid Aga, Apsara Sivaprazad, Mohamed Al-Musleh

机构 * Physical Sciences Heriot Watt University Dubai, UAE（物理科学 哈罗伊特沃特大学 迪拜, 英国） ； Computer Sciences Heriot Watt University Dubai, UAE（计算机科学 哈罗伊特沃特大学 迪拜, 英国） 纠错

专题命中 规划推理 ：planning(abstract)

Euclid's Gift: Enhancing Spatial Perception and Reasoning in Vision-Language Models via Geometric Surrogate Tasks

Shijie Lian, Changti Wu, Laurence Tianruo Yang, Hang Yuan, Bin Yu, Lei Zhang, Kai Chen

机构 * Huazhong University of Science and Technology（华中科技大学） ； Zhongguancun Academy（中关村学院） ； East China Normal University（华东师范大学） ； Zhengzhou University（郑州大学） ； Zhongguancun Institute of Artificial Intelligence（中关村人工智能研究院） 纠错

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

Graph Query Networks for Object Detection with Automotive Radar

Loveneet Saini, Hasan Tercan, Tobias Meisen

机构 * University of Wuppertal（乌珀塔尔大学） 纠错

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

Comments Accepted in WACV 2026 Main Conference

Eq.Bot: Enhance Robotic Manipulation Learning via Group Equivariant Canonicalization

Jian Deng, Yuandong Wang, Yangfu Zhu, Tao Feng, Tianyu Wo, Zhenzhou Shao

机构 * Beijing Key Laboratory of Light Industrial Robot and Safety Verification, Capital Normal University（北京工业机器人与安全验证重点实验室，首都师范大学） ； Department of Computer Science and Technology, Tsinghua University（清华大学计算机科学与技术系） ； School of Software, Beihang University（北航软件学院） 纠错

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

Comments 12 pages, 4 figures and 3 tables

One algebra for all : Geometric Algebra methods for neurosymbolic XR scene authoring, animation and neural rendering

Manos Kamarianakis, Antonis Protopsaltis, George Papagiannakis

专题命中 视觉空间推理 ：reasoning(abstract)

Comments 10 pages, 9 Figures

C2F-Space: Coarse-to-Fine Space Grounding for Spatial Instructions using Vision-Language Models

Nayoung Oh, Dohyun Kim, Junhyeong Bang, Rohan Paul, Daehyung Park

专题命中 视觉空间推理 ：reasoning(abstract)

Comments 16 pages, 12 figures

Temporal Predictors of Outcome in Reasoning Language Models

Joey David

机构 * Independent Researcher（独立研究者） 纠错

专题命中 测试时计算 ：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.CL

Comments 4 pages, 4 figures

Trade-offs in Large Reasoning Models: An Empirical Analysis of Deliberative and Adaptive Reasoning over Foundational Capabilities

Weixiang Zhao, Xingyu Sui, Jiahe Guo, Yulin Hu, Yang Deng, Yanyan Zhao, Xuda Zhi, Yongbo Huang, Hao He, Wanxiang Che, Ting Liu, Bing Qin

专题命中 测试时计算 ：reasoning(title,abstract)；chain-of-thought(abstract)；分类 cs.CL、cs.AI

Comments To appear at AAAI 2026

Foundational Automatic Evaluators: Scaling Multi-Task Generative Evaluator Training for Reasoning-Centric Domains

Austin Xu, Xuan-Phi Nguyen, Yilun Zhou, Chien-Sheng Wu, Caiming Xiong, Shafiq Joty

机构 * Salesforce AI Research（Salesforce人工智能研究） 纠错

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

Comments 29 pages, 9 tables, 6 figures

Think Smart, Not Hard: Difficulty Adaptive Reasoning for Large Audio Language Models

Zhichao Sheng, Shilin Zhou, Chen Gong, Zhenghua Li

专题命中 复杂问题求解 ：reasoning(title,abstract)；chain-of-thought(abstract)；CoT(abstract)；分类 cs.LG

Efficiency Will Not Lead to Sustainable Reasoning AI

Philipp Wiesner, Daniel W. O'Neill, Francesca Larosa, Odej Kao

机构 * TU Berlin（柏林技术大学） ； Universitat de Barcelona（巴塞罗那大学） ； KTH Royal Institute of Technology（皇家理工学院） 纠错

专题命中 复杂问题求解 ：reasoning(title,abstract)；分类 cs.AI

Comments Presented at the Rethinking AI Workshop @ EurIPS'25

S-DAG: A Subject-Based Directed Acyclic Graph for Multi-Agent Heterogeneous Reasoning

Jiangwen Dong, Zehui Lin, Wanyu Lin, Mingjin Zhang

机构 * PolyU 纠错

专题命中 复杂问题求解 ：reasoning(title,abstract)；分类 cs.AI

Comments Accepted by AAAI 2026

Reasoning in Diffusion Large Language Models is Concentrated in Dynamic Confusion Zones

Ranfei Chen, Ming Chen, Kaifei Wang

机构 * Institute of Computing Technology Chinese Academy of Sciences Beijing, China（计算技术研究所中国科学院北京） 纠错

专题命中 复杂问题求解 ：reasoning(title,abstract)；分类 cs.LG