语言大模型 / LLM - arXivDaily 专题

2506.17088 2025-09-17 cs.CL 92%

Chain-of-Thought Prompting Obscures Hallucination Cues in Large Language Models: An Empirical Evaluation

Jiahao Cheng, Tiancheng Su, Jia Yuan, Guoxiu He, Jiawei Liu, Xinqi Tao, Jingwen Xie, Huaxia Li

机构 * East China Normal University（东中国师范大学）； Wuhan University（武汉大学）； Xiaohongshu Inc.（小红书公司）

专题命中推理与问题求解：large language model(title,abstract)；language model(title,abstract)；prompting(title,abstract)；LLM(abstract)

Comments Accepted at EMNLP 2025 Findings

2509.12645 2025-09-17 cs.AI cs.LO 89%

Large Language Models Imitate Logical Reasoning, but at what Cost?

Lachlan McGinness, Peter Baumgartner

机构 * Australian National University（澳大利亚国立大学）； Data61|CSIRO

专题命中推理与问题求解：large language model(title,abstract)；language model(title,abstract)；prompting(abstract)；分类 cs.AI

Comments This work has been accepted as a main track paper for publication in the proceedings of the Australasian Joint Conference on Artificial Intelligence 2025 held in Canberra, Australia

2506.08171 2025-09-17 cs.SE cs.AI 89%

Worst-Case Symbolic Constraints Analysis and Generalisation with Large Language Models

Daniel Koh, Yannic Noller, Corina S. Pasareanu, Adrians Skapars, Youcheng Sun

机构 * Mohamed bin Zayed University of Artificial Intelligence（莫扎德·本·扎耶德人工智能大学）； Carnegie Mellon University（卡内基梅隆大学）； University of Manchester（曼彻斯特大学）

专题命中推理与问题求解：large language model(title,abstract)；language model(title,abstract)；LLM(abstract)；分类 cs.AI

2507.08679 2025-09-17 cs.CV 89%

ByDeWay: Boost Your multimodal LLM with DEpth prompting in a Training-Free Way

Rajarshi Roy, Devleena Das, Ankesh Banerjee, Arjya Bhattacharjee, Kousik Dasgupta, Subarna Tripathi

机构 * Kalyani Government Engineering College（卡利尼政府工程学院）； Intel Labs（英特尔实验室）

专题命中推理与问题求解：prompting(title,abstract)；LLM(title)；large language model(abstract)；language model(abstract)

2509.12908 2025-09-17 cs.CL cs.AI 88%

All Roads Lead to Rome: Graph-Based Confidence Estimation for Large Language Model Reasoning

Caiqi Zhang, Chang Shu, Ehsan Shareghi, Nigel Collier

机构 * University of Cambridge（剑桥大学）； Monash University（墨尔本大学）

专题命中推理与问题求解：large language model(title,abstract)；language model(title,abstract)；分类 cs.CL、cs.AI

Comments EMNLP 2025 Main

2509.12060 2025-09-17 cs.AI 88%

When Safe Unimodal Inputs Collide: Optimizing Reasoning Chains for Cross-Modal Safety in Multimodal Large Language Models

Wei Cai, Shujuan Liu, Jian Zhao, Ziyan Shi, Yusheng Zhao, Yuchen Yuan, Tianle Zhang, Chi Zhang, Xuelong Li

专题命中推理与问题求解：large language model(title,abstract)；language model(title,abstract)；分类 cs.AI

2507.14975 2025-09-17 cs.RO cs.AI 88%

FCRF: Flexible Constructivism Reflection for Long-Horizon Robotic Task Planning with Large Language Models

Yufan Song, Jiatao Zhang, Zeng Gu, Qingmiao Liang, Tuocheng Hu, Wei Song, Shiqiang Zhu

机构 * Zhejiang University（浙江大学）； Hangzhou Institute for Advanced Study（杭州高等研究 institute）

专题命中推理与问题求解：large language model(title,abstract)；language model(title,abstract)；分类 cs.AI

Comments 8 pages, 6 figures, IROS 2025

2509.13237 2025-09-17 cs.LG cs.AI 88%

Metacognitive Reuse: Turning Recurring LLM Reasoning Into Concise Behaviors

Aniket Didolkar, Nicolas Ballas, Sanjeev Arora, Anirudh Goyal

机构 * Meta ； Mila-Quebec AI Institute, University of Montreal（魁北克AI研究所，蒙特利尔大学）； Princeton University（普林斯顿大学）； Mila--Quebec AI Institute, Université de Montréal（魁北克AI研究所，蒙特利尔大学）

专题命中推理与问题求解：LLM(title,abstract)；large language model(abstract)；language model(abstract)；SFT(abstract)

Comments 18 pages, 9 Figures, 5 Tables

2509.12273 2025-09-17 cs.AI cs.CL cs.LG 87%

LLMAP: LLM-Assisted Multi-Objective Route Planning with User Preferences

Liangqi Yuan, Dong-Jun Han, Christopher G. Brinton, Sabine Brunswicker

专题命中推理与问题求解：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI、cs.LG

2509.13131 2025-09-17 cs.AI 85%

Reasoning with Preference Constraints: A Benchmark for Language Models in Many-to-One Matching Markets

Marylou Fauchard, Florian Carichon, Margarida Carvalho, Golnoosh Farnadi

机构 * Université de Montréal（蒙特利尔大学）； Mila ； McGill（麦吉尔大学）

专题命中推理与问题求解：language model(title,abstract)；large language model(abstract)；prompting(abstract)；分类 cs.AI

2509.12611 2025-09-17 cs.AI 85%

Analogy-Driven Financial Chain-of-Thought (AD-FCoT): A Prompting Approach for Financial Sentiment Analysis

Anmol Singhal Navya Singhal

机构 * New York University（纽约大学）； University of Texas at Austin（德克萨斯大学奥斯汀分校）

专题命中推理与问题求解：prompting(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI

Comments IEEE AIxB 2025

2406.15444 2025-09-17 cs.CL 85%

Cutting Through the Noise: Boosting LLM Performance on Math Word Problems

Ujjwala Anantheswaran, Himanshu Gupta, Kevin Scaria, Shreyas Verma, Chitta Baral, Swaroop Mishra

机构 * Arizona State University（亚利桑那州立大学）； Georgia Institute of Technology（佐治亚理工学院）

专题命中推理与问题求解：LLM(title)；large language model(abstract)；language model(abstract)；prompting(abstract)

Comments Published at ICLR 2025 Workshop on Reasoning and Planning for LLMs

2509.12638 2025-09-17 cs.CE 85%

FinSentLLM: Multi-LLM and Structured Semantic Signals for Enhanced Financial Sentiment Forecasting

Zijian Zhang, Rong Fu, Yangfan He, Xinze Shen, Yanlong Wang, Xiaojing Du, Haochen You, Jiazhao Shi, Simon Fong

专题命中推理与问题求解：LLM(title,abstract)；large language model(abstract)；language model(abstract)

2507.09937 2025-09-17 cs.LG cs.AI 84%

Memorization Sinks: Isolating Memorization during LLM Training

Gaurav R. Ghosal, Pratyush Maini, Aditi Raghunathan

专题命中推理与问题求解：LLM(title)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Comments Accepted at the 2025 International Conference of Machine Learning

2509.12914 2025-09-17 cs.AI 83%

Stochastic Streets: A Walk Through Random LLM Address Generation in four European Cities

Tairan Fu, David Campo-Nazareno, Javier Coronado-Blázquez, Javier Conde, Pedro Reviriego, Fabrizio Lombardi

专题命中推理与问题求解：LLM(title)；large language model(abstract)；language model(abstract)；分类 cs.AI

2509.12798 2025-09-17 cs.SE cs.AI 83%

LLM-Based Approach for Enhancing Maintainability of Automotive Architectures

Nenad Petrovic, Lukasz Mazur, Alois Knoll

机构 * Technical University of Munich（慕尼黑技术大学）

专题命中推理与问题求解：LLM(title)；large language model(abstract)；language model(abstract)；分类 cs.AI

2505.14172 2025-09-17 cs.CL 83%

The Strawberry Problem: Emergence of Character-level Understanding in Tokenized Language Models

Adrian Cosma, Stefan Ruseti, Emilian Radoi, Mihai Dascalu

专题命中推理与问题求解：language model(title,abstract)；large language model(abstract)；分类 cs.CL

Comments Accepted at EMNLP 2025 Main as Oral Presentation (Top 15% of accepted papers)

2509.13317 2025-09-17 cs.CV 82%

3D Aware Region Prompted Vision Language Model

An-Chieh Cheng, Yang Fu, Yukang Chen, Zhijian Liu, Xiaolong Li, Subhashree Radhakrishnan, Song Han, Yao Lu, Jan Kautz, Pavlo Molchanov, Hongxu Yin, Xiaolong Wang, Sifei Liu

机构 * MIT（麻省理工学院）； NVIDIA（英伟达）

专题命中推理与问题求解：language model(title,abstract)；prompting(abstract)

Comments Project Website: https://www.anjiecheng.me/sr3d

2509.12743 2025-09-17 cs.AI cs.CL 79%

Zero-shot Graph Reasoning via Retrieval Augmented Framework with LLMs

Hanqing Li, Kiran Sheena Jyothi, Henry Liang, Sharika Mahadevan, Diego Klabjan

机构 * Northwestern University（西北大学）； EXL Service（EXL服务）； Vail Systems（Vail系统）； Netflix

专题命中推理与问题求解：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2502.12067 2025-09-17 cs.CL cs.AI 79%

TokenSkip: Controllable Chain-of-Thought Compression in LLMs

Heming Xia, Chak Tou Leong, Wenjie Wang, Yongqi Li, Wenjie Li

机构 * Department of Computing, The Hong Kong Polytechnic University（计算学系，香港理工大学）

专题命中推理与问题求解：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments EMNLP 2025 (Long Paper), camera-ready version

2509.13127 2025-09-17 cs.CL 77%

Empowering LLMs with Parameterized Skills for Adversarial Long-Horizon Planning

Sijia Cui, Shuai Xu, Aiyao He, Yanna Wang, Bo Xu

机构 * 1 Institute of Automation, Chinese Academy of Sciences, Beijing, China ； 2 School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China ； 3 Nanjing Artificial Intelligence Research of IA, Nanjing, China ； 4 University of Chinese Academy of Sciences,Nanjing, Nanjing, China ； 5 Nanjing University of Information Science \& Technology, Nanjing, China

专题命中推理与问题求解：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL

Comments Accepted to IJCNN 2025

2509.12987 2025-09-17 cs.AI 77%

Toward PDDL Planning Copilot

Yarin Benyamin, Argaman Mordoch, Shahaf S. Shperberg, Roni Stern

专题命中推理与问题求解：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI

2509.12720 2025-09-17 cs.CL 77%

HistoryBankQA: Multilingual Temporal Question Answering on Historical Events

Biswadip Mandal, Anant Khandelwal, Manish Gupta

专题命中推理与问题求解：large language model(abstract)；language model(abstract)；small language model(abstract)；分类 cs.CL

2509.12603 2025-09-17 cs.CL cs.AI 73%

EconProver: Towards More Economical Test-Time Scaling for Automated Theorem Proving

Mukai Li, Linfeng Song, Zhenwen Liang, Jiahao Xu, Shansan Gong, Qi Liu, Haitao Mi, Dong Yu

机构 * Tencent（腾讯）； The University of Hong Kong（香港大学）

专题命中推理与问题求解：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

2410.10596 2025-09-17 cs.AI cs.LG q-bio.NC 73%

Overcoming classic challenges for artificial neural networks by providing incentives and practice

Kazuki Irie, Brenden M. Lake

专题命中推理与问题求解：large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Comments In press at Nature Machine Intelligence

2509.13282 2025-09-17 cs.CL cs.CV cs.LG 62%

ChartGaze: Enhancing Chart Understanding in LVLMs with Eye-Tracking Guided Attention Refinement

Ali Salamatian, Amirhossein Abaskohi, Wan-Cyuan Fan, Mir Rayat Imtiaz Hossain, Leonid Sigal, Giuseppe Carenini

专题命中推理与问题求解：language model(abstract)；分类 cs.CL、cs.LG

Comments EMNLP 2025

2509.12371 2025-09-17 cs.CL cs.AI 62%

MORABLES: A Benchmark for Assessing Abstract Moral Reasoning in LLMs with Fables

Matteo Marcuzzo, Alessandro Zangari, Andrea Albarelli, Jose Camacho-Collados, Mohammad Taher Pilehvar

专题命中推理与问题求解：LLM(abstract)；分类 cs.CL、cs.AI

Comments Accepted to EMNLP 2025 Main Conference

2509.12233 2025-09-17 cs.CR cs.AI cs.ET cs.LG cs.NI 62%

Towards Trustworthy Agentic IoEV: AI Agents for Explainable Cyberthreat Mitigation and State Analytics

Meryem Malak Dif, Mouhamed Amine Bouchiha, Abdelaziz Amara Korba, Yacine Ghamri-Doudane

机构 * L3i - La Rochelle University, La Rochelle, France（L3i - 拉罗谢尔大学）

专题命中推理与问题求解：LLM(abstract)；分类 cs.AI、cs.LG

Comments 10 pages, 7 figures, Accepted at LCN'25

2509.05066 2025-09-17 cs.CL cs.AI 62%

ToM-SSI: Evaluating Theory of Mind in Situated Social Interactions

Matteo Bortoletto, Constantin Ruhdorfer, Andreas Bulling

专题命中推理与问题求解：foundation model(abstract)；分类 cs.CL、cs.AI

Comments EMNLP 2025 (Main)

2509.12589 2025-09-17 cs.AI 57%

Redefining CX with Agentic AI: Minerva CQ Case Study

Garima Agrawal, Riccardo De Maria, Kiran Davuluri, Daniele Spera, Charlie Read, Cosimo Spera, Jack Garrett, Don Miller

机构 * Minerva CQ AI Research Division（Minerva CQ人工智能研究部）

专题命中推理与问题求解：prompting(abstract)；分类 cs.AI

1. 推理与问题求解 32 篇

Chain-of-Thought Prompting Obscures Hallucination Cues in Large Language Models: An Empirical Evaluation

Large Language Models Imitate Logical Reasoning, but at what Cost?

Worst-Case Symbolic Constraints Analysis and Generalisation with Large Language Models

ByDeWay: Boost Your multimodal LLM with DEpth prompting in a Training-Free Way

All Roads Lead to Rome: Graph-Based Confidence Estimation for Large Language Model Reasoning

When Safe Unimodal Inputs Collide: Optimizing Reasoning Chains for Cross-Modal Safety in Multimodal Large Language Models

FCRF: Flexible Constructivism Reflection for Long-Horizon Robotic Task Planning with Large Language Models

Metacognitive Reuse: Turning Recurring LLM Reasoning Into Concise Behaviors

LLMAP: LLM-Assisted Multi-Objective Route Planning with User Preferences

Reasoning with Preference Constraints: A Benchmark for Language Models in Many-to-One Matching Markets

Analogy-Driven Financial Chain-of-Thought (AD-FCoT): A Prompting Approach for Financial Sentiment Analysis

Cutting Through the Noise: Boosting LLM Performance on Math Word Problems

FinSentLLM: Multi-LLM and Structured Semantic Signals for Enhanced Financial Sentiment Forecasting

Memorization Sinks: Isolating Memorization during LLM Training

Stochastic Streets: A Walk Through Random LLM Address Generation in four European Cities

LLM-Based Approach for Enhancing Maintainability of Automotive Architectures

The Strawberry Problem: Emergence of Character-level Understanding in Tokenized Language Models

3D Aware Region Prompted Vision Language Model

Zero-shot Graph Reasoning via Retrieval Augmented Framework with LLMs

TokenSkip: Controllable Chain-of-Thought Compression in LLMs

Empowering LLMs with Parameterized Skills for Adversarial Long-Horizon Planning

Toward PDDL Planning Copilot

HistoryBankQA: Multilingual Temporal Question Answering on Historical Events

EconProver: Towards More Economical Test-Time Scaling for Automated Theorem Proving

Overcoming classic challenges for artificial neural networks by providing incentives and practice

ChartGaze: Enhancing Chart Understanding in LVLMs with Eye-Tracking Guided Attention Refinement

MORABLES: A Benchmark for Assessing Abstract Moral Reasoning in LLMs with Fables

Towards Trustworthy Agentic IoEV: AI Agents for Explainable Cyberthreat Mitigation and State Analytics

ToM-SSI: Evaluating Theory of Mind in Situated Social Interactions

Redefining CX with Agentic AI: Minerva CQ Case Study