大模型推理能力 - arXivDaily 专题

2510.03514 2025-10-07 cs.CY cs.AI cs.CL 62%

Red Lines and Grey Zones in the Fog of War: Benchmarking Legal Risk, Moral Harm, and Regional Bias in Large Language Model Military Decision-Making

Toby Drinkall

机构 * Oxford Internet Institute, University of Oxford（牛津互联网研究所、牛津大学）

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

Comments 54 pages; 11 figures

2510.03065 2025-10-06 cs.LG cs.AI 62%

A Unified Deep Reinforcement Learning Approach for Close Enough Traveling Salesman Problem

Mingfeng Fan, Jiaqi Cheng, Yaoxin Wu, Yifeng Zhang, Yibin Yang, Guohua Wu, Guillaume Sartoretti

机构 * Department of Mechanical Engineering, National University of Singapore（新加坡国立大学机械工程系）； School of Traffic and Transportation Engineering, Central South University（中南大学交通与运输工程学院）； Department of Industrial Engineering and Innovation Sciences, Eindhoven University of Technology（埃因霍温理工大学工业工程与创新科学系）； School of Vehicle and Mobility, Tsinghua University（清华大学车辆与移动性学院）； School of Automation, Central South University（中南大学自动化学院）

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

2510.02084 2025-10-06 cs.LG cs.AI 62%

KAIROS: Unified Training for Universal Non-Autoregressive Time Series Forecasting

Kuiye Ding, Fanda Fan, Zheya Wang, Hongxiao Li, Yifan Wang, Lei Wang, Chunjie Luo, Jianfeng Zhan

机构 * Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）； Department of Mathematical Sciences, Durham University（杜伦大学数学科学系）； University of Chinese Academy of Sciences（中国科学院大学）

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

2510.00038 2025-10-06 cs.LG cs.AI cs.CY 62%

DM-Bench: Benchmarking LLMs for Personalized Decision Making in Diabetes Management

Maria Ana Cardei, Josephine Lamp, Mark Derdzinski, Karan Bhatia

机构 * University of Virginia（弗吉尼亚大学）； Dexcom（德科姆公司）

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

2503.07044 2025-10-06 cs.CL cs.AI 62%

DatawiseAgent: A Notebook-Centric LLM Agent Framework for Adaptive and Robust Data Science Automation

Ziming You, Yumiao Zhang, Dexuan Xu, Yiwei Lou, Yandong Yan, Wei Wang, Huaming Zhang, Yu Huang

机构 * National Engineering Research Center for Software Engineering, Peking University（软件工程国家工程研究中心，北京大学）； School of Software & Microelectronics, Peking University（软件与微电子学院，北京大学）； School of Computer Science, Peking University（计算机科学学院，北京大学）； Xi’an Jiaotong University（西安交通大学）； Institute of Basic Theory of Chinese Medicine, China Academy of Chinese Medical Sciences（中医基础理论研究所，中国中医科学院）

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

Comments The camera-ready version for EMNLP 2025 Main Conference

2510.01531 2025-10-03 cs.AI cs.CL cs.RO 62%

Information Seeking for Robust Decision Making under Partial Observability

Djengo Cyun-Jyun Fang, Tsung-Wei Ke

机构 * National Taiwan University（台湾大学）

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

Comments The project page is available at https://infoseekerllm.github.io

2510.01524 2025-10-03 cs.CV cs.AI cs.LG 62%

WALT: Web Agents that Learn Tools

Viraj Prabhu, Yutong Dai, Matthew Fernandez, Jing Gu, Krithika Ramakrishnan, Yanqi Luo, Silvio Savarese, Caiming Xiong, Junnan Li, Zeyuan Chen, Ran Xu

机构 * Salesforce AI Research（Salesforce AI研究院）

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

2510.01279 2025-10-03 cs.CL cs.AI 62%

TUMIX: Multi-Agent Test-Time Scaling with Tool-Use Mixture

Yongchao Chen, Jiefeng Chen, Rui Meng, Ji Yin, Na Li, Chuchu Fan, Chi Wang, Tomas Pfister, Jinsung Yoon

机构 * MIT Harvard（麻省理工学院哈佛大学）； Google Cloud AI Research（谷歌云人工智能研究）； Google DeepMind（谷歌DeepMind）； MIT（麻省理工学院）

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

Comments 27 pages, 13 figures

2510.01229 2025-10-03 cs.CL cs.AI 62%

Enhancing Transformer-Based Rerankers with Synthetic Data and LLM-Based Supervision

Dimitar Peshevski, Kiril Blazhevski, Martin Popovski, Gjorgji Madjarov

机构 * Faculty of Computer Science and Engineering（计算机科学与工程学院）； Machine Learning Department（机器学习系）

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

Comments Accepted by RANLP 2025

2510.01226 2025-10-03 cs.CL cs.AI 62%

ClaimCheck: Real-Time Fact-Checking with Small Language Models

Akshith Reddy Putta, Jacob Devasier, Chengkai Li

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

2505.14620 2025-10-03 cs.LG cs.CL 62%

Enhancing Learned Knowledge in LoRA Adapters Through Efficient Contrastive Decoding on Ascend NPUs

Morgan Lindsay Heisler, Linzi Xing, Ge Shi, Hanieh Sadri, Gursimran Singh, Weiwei Zhang, Tao Ye, Ying Xiong, Yong Zhang, Zhenan Fan

机构 * Huawei Technologies Canada（华为技术加拿大公司）； Huawei Technologies（华为技术）

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

Comments Accepted at ACM KDD 2025

2510.00452 2025-10-02 cs.CR cs.AI cs.LG cs.MA 62%

Cloud Investigation Automation Framework (CIAF): An AI-Driven Approach to Cloud Forensics

Dalal Alharthi, Ivan Roberto Kawaminami Garcia

机构 * University of Arizona（亚利桑那大学）

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

2510.00401 2025-10-02 cs.RO cs.AI cs.LG cs.MA 62%

Physics-Informed Neural Controlled Differential Equations for Scalable Long Horizon Multi-Agent Motion Forecasting

Shounak Sural, Charles Kekeh, Wenliang Liu, Federico Pecora, Mouhacine Benosman

机构 * Carnegie Mellon University（卡内基梅隆大学）； Amazon Robotics（亚马逊机器人）

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

2510.00078 2025-10-02 cs.LG cs.AI cs.DC 62%

Adaptive and Resource-efficient Agentic AI Systems for Mobile and Embedded Devices: A Survey

Sicong Liu, Weiye Wu, Xiangrui Xu, Teng Li, Bowen Pang, Bin Guo, Zhiwen Yu

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

2502.19130 2025-10-01 cs.MA cs.AI cs.CL 62%

Voting or Consensus? Decision-Making in Multi-Agent Debate

Lars Benedikt Kaesberg, Jonas Becker, Jan Philip Wahle, Terry Ruas, Bela Gipp

机构 * University of Göttingen（戈丁根大学）； LKA NRW（NRW研究所）

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

Comments Accepted at ACL2025 (Findings)

Journal ref ACL 2025

2407.11843 2025-10-01 cs.CL cs.AI 62%

Preemptive Detection and Correction of Misaligned Actions in LLM Agents

Haishuo Fang, Xiaodan Zhu, Iryna Gurevych

机构 * Ubiquitous Knowledge Processing Lab, Department of Computer Science, TU Darmstadt（图宾根大学计算机科学系）； National Research Center for Applied Cybersecurity ATHENE, Germany（德国应用网络安全国家研究中心ATHENE）； Department of Electrical and Computer Engineering & Ingenuity Labs Research Institute, Queen’s University, Canada（加拿大女王大学电气与计算机工程系及创新实验室研究院）

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

Comments Accepted by EMNLP 2025

2509.25189 2025-09-30 cs.CL cs.AI 62%

InfoAgent: Advancing Autonomous Information-Seeking Agents

Gongrui Zhang, Jialiang Zhu, Ruiqi Yang, Kai Qiu, Miaosen Zhang, Zhirong Wu, Qi Dai, Bei Liu, Chong Luo, Zhengyuan Yang, Linjie Li, Lijuan Wang, Weizhu Chen, Yuan Zhang, Xin Li, Zhaoyi Liu, Xin Geng, Baining Guo

机构 * Southeast University（东南大学）； Brown University（布朗大学）； Microsoft（微软）

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

2509.24107 2025-09-30 cs.AI cs.LG 62%

Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs

Shreyas Singh, Kunal Singh, Pradeep Moturi

机构 * Fractal AI Research（Fractal AI研究）

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

2509.22921 2025-09-30 cs.LG cs.AI 62%

Rethinking Large Language Model Distillation: A Constrained Markov Decision Process Perspective

Matthieu Zimmer, Xiaotong Ji, Tu Nguyen, Haitham Bou Ammar

机构 * Huawei Noah’s Ark Lab（华为诺亚实验室）； Huawei R&D Munich（华为慕尼黑研发部）； UCL Centre for Artificial Intelligence（伦敦大学学院人工智能中心）

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

2509.22130 2025-09-29 cs.MA cs.AI cs.LG 62%

Multi-Agent Path Finding via Offline RL and LLM Collaboration

Merve Atasever, Matthew Hong, Mihir Nitin Kulkarni, Qingpei Li, Jyotirmoy V. Deshmukh

机构 * Department of Computer Science, University of Southern California（计算机科学系，南加州大学）

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

2508.00719 2025-09-29 cs.CL cs.AI 62%

DAMR: Efficient and Adaptive Context-Aware Knowledge Graph Question Answering with LLM-Guided MCTS

Yingxu Wang, Shiqi Fan, Mengzhu Wang, Siyang Gao, Chao Wang, Nan Yin

机构 * MBZUAI ； PolyU ； Hebei University of Technology（河北工业大学）； CityU（城市大学）； HKUST（香港科技大学）

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

2505.23537 2025-09-29 cs.LG cs.CL 62%

Domain-Aware Tensor Network Structure Search

Giorgos Iacovides, Wuyang Zhou, Chao Li, Qibin Zhao, Danilo Mandic

机构 * Department of Electrical and Electronic Engineering, Imperial College London（帝国理工学院伦敦校区电子与电气工程系）； RIKEN AIP（日本理化学研究所AIP）

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

2509.19657 2025-09-25 cs.CL cs.AI cs.SI 62%

Large Language Models for Pedestrian Safety: An Application to Predicting Driver Yielding Behavior at Unsignalized Intersections

Yicheng Yang, Zixian Li, Jean Paul Bizimana, Niaz Zafri, Yongfeng Dong, Tianyi Li

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

2509.19369 2025-09-25 cs.CL cs.AI 62%

SLM-Based Agentic AI with P-C-G: Optimized for Korean Tool Use

Changhyun Jeon, Jinhee Park, Jungwoo Choi, Keonwoo Kim, Jisu Kim, Minji Hong

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

2509.18141 2025-09-24 cs.LG cs.AI cs.CV stat.AP stat.ML 62%

KM-GPT: An Automated Pipeline for Reconstructing Individual Patient Data from Kaplan-Meier Plots

Yao Zhao, Haoyue Sun, Yantian Ding, Yanxun Xu

机构 * Johns Hopkins University（约翰霍普金斯大学）

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

2411.16120 2025-09-23 cs.AI cs.LG 62%

Why the Agent Made that Decision: Contrastive Explanation Learning for Reinforcement Learning

Rui Zuo, Simon Khan, Zifan Wang, Garrett Ethan Katz, Qinru Qiu

机构 * Syracuse University（苏塞克斯大学）； Air Force Research Laboratory（空军研究实验室）

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

2509.17165 2025-09-23 cs.LG cs.AI 62%

Time Series Forecasting Using a Hybrid Deep Learning Method: A Bi-LSTM Embedding Denoising Auto Encoder Transformer

Sahar Koohfar, Wubeshet Woldemariam

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

2505.17616 2025-09-23 cs.CL cs.AI 62%

Runaway is Ashamed, But Helpful: On the Early-Exit Behavior of Large Language Model-based Agents in Embodied Environments

Qingyu Lu, Liang Ding, Siyi Cao, Xuebo Liu, Kanjian Zhang, Jinxia Zhang, Dacheng Tao

机构 * Southeast University（东南大学）； The University of Sydney（悉尼大学）； Southeast University Shenzhen Research Institute（东南大学深圳研究院）； Institute of Computing and Intelligence, Harbin Institute of Technology, Shenzhen, China（哈尔滨工业大学计算机与智能信息学院，深圳，中国）； College of Computing and Data Science at Nanyang Technological University, Singapore 639798（南洋理工大学计算机与数据科学学院，新加坡）

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

Comments EMNLP 2025 - Findings

2509.15915 2025-09-22 cs.LG cs.AI 62%

Foundation Models as World Models: A Foundational Study in Text-Based GridWorlds

Remo Sasso, Michelangelo Conserva, Dominik Jeurissen, Paulo Rauber

机构 * School of Electronic Engineering and Computer Science（电子工程与计算机科学学院）

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

Comments 20 pages, 9 figures. Accepted for presentation at the 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop on Embodied World Models for Decision Making

2503.12123 2025-09-22 cs.CL cs.AI 62%

MT-RewardTree: A Comprehensive Framework for Advancing LLM-Based Machine Translation via Reward Modeling

Zhaopeng Feng, Jiahan Ren, Jiayuan Su, Jiamei Zheng, Hongwei Wang, Zuozhu Liu

机构 * Zhejiang University（浙江大学）

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

Comments EMNLP 2025 Findings. Project page:https://sabijun.github.io/MT_RewardTreePage

1. 规划推理 11104 篇

Red Lines and Grey Zones in the Fog of War: Benchmarking Legal Risk, Moral Harm, and Regional Bias in Large Language Model Military Decision-Making

A Unified Deep Reinforcement Learning Approach for Close Enough Traveling Salesman Problem

KAIROS: Unified Training for Universal Non-Autoregressive Time Series Forecasting

DM-Bench: Benchmarking LLMs for Personalized Decision Making in Diabetes Management

DatawiseAgent: A Notebook-Centric LLM Agent Framework for Adaptive and Robust Data Science Automation

Information Seeking for Robust Decision Making under Partial Observability

WALT: Web Agents that Learn Tools

TUMIX: Multi-Agent Test-Time Scaling with Tool-Use Mixture

Enhancing Transformer-Based Rerankers with Synthetic Data and LLM-Based Supervision

ClaimCheck: Real-Time Fact-Checking with Small Language Models

Enhancing Learned Knowledge in LoRA Adapters Through Efficient Contrastive Decoding on Ascend NPUs

Cloud Investigation Automation Framework (CIAF): An AI-Driven Approach to Cloud Forensics

Physics-Informed Neural Controlled Differential Equations for Scalable Long Horizon Multi-Agent Motion Forecasting

Adaptive and Resource-efficient Agentic AI Systems for Mobile and Embedded Devices: A Survey

Voting or Consensus? Decision-Making in Multi-Agent Debate

Preemptive Detection and Correction of Misaligned Actions in LLM Agents

InfoAgent: Advancing Autonomous Information-Seeking Agents

Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs

Rethinking Large Language Model Distillation: A Constrained Markov Decision Process Perspective

Multi-Agent Path Finding via Offline RL and LLM Collaboration

DAMR: Efficient and Adaptive Context-Aware Knowledge Graph Question Answering with LLM-Guided MCTS

Domain-Aware Tensor Network Structure Search

Large Language Models for Pedestrian Safety: An Application to Predicting Driver Yielding Behavior at Unsignalized Intersections

SLM-Based Agentic AI with P-C-G: Optimized for Korean Tool Use

KM-GPT: An Automated Pipeline for Reconstructing Individual Patient Data from Kaplan-Meier Plots

Why the Agent Made that Decision: Contrastive Explanation Learning for Reinforcement Learning

Time Series Forecasting Using a Hybrid Deep Learning Method: A Bi-LSTM Embedding Denoising Auto Encoder Transformer

Runaway is Ashamed, But Helpful: On the Early-Exit Behavior of Large Language Model-based Agents in Embodied Environments

Foundation Models as World Models: A Foundational Study in Text-Based GridWorlds

MT-RewardTree: A Comprehensive Framework for Advancing LLM-Based Machine Translation via Reward Modeling