大模型对齐与安全 - arXivDaily 专题

2509.23457 2025-09-30 cs.CV 50%

No Concept Left Behind: Test-Time Optimization for Compositional Text-to-Image Generation

Mohammad Hossein Sameti, Amir M. Mansourian, Arash Marioriyad, Soheil Fadaee Oshyani, Mohammad Hossein Rohban, Mahdieh Soleymani Baghshah

机构 * Sharif University of Technology（沙里夫技术大学）

专题命中幻觉与事实性：alignment(abstract)

Comments 8 pages, 8 figures, 1 table

2509.23425 2025-09-30 cs.MA 50%

Situational Awareness for Safe and Robust Multi-Agent Interactions Under Uncertainty

Benjamin Alcorn, Eman Hammad

专题命中幻觉与事实性：safety(abstract)

2504.11669 2025-09-23 cs.CV 50%

Co-STAR: Collaborative Curriculum Self-Training with Adaptive Regularization for Source-Free Video Domain Adaptation

Amirhossein Dadashzadeh, Parsa Esmati, Majid Mirmehdi

机构 * University of Bristol, UK（布里斯托大学）

专题命中幻觉与事实性：alignment(abstract)

2509.14084 2025-09-19 cs.CV 50%

AD-DINOv3: Enhancing DINOv3 for Zero-Shot Anomaly Detection with Anomaly-Aware Calibration

Jingyi Yuan, Jianxiong Ye, Wenkang Chen, Chenqiang Gao

机构 * School of Intelligent Systems Engineering, Sun Yat-Sen University（智能系统工程学院，中山大学）

专题命中幻觉与事实性：alignment(abstract)

2509.09206 2025-09-18 cs.RO cs.SY eess.SY 50%

Occupancy-aware Trajectory Planning for Autonomous Valet Parking in Uncertain Dynamic Environments

Farhad Nawaz, Faizan M. Tariq, Sangjae Bae, David Isele, Avinash Singh, Nadia Figueroa, Nikolai Matni, Jovin D'sa

机构 * Honda Research Institute (HRI)（本田研究院）； University of Pennsylvania（宾夕法尼亚大学）

专题命中幻觉与事实性：safety(abstract)

2509.09989 2025-09-15 cs.CR 50%

rCamInspector: Building Reliability and Trust on IoT (Spy) Camera Detection using XAI

Priyanka Rushikesh Chaudhary, Manan Gupta, Jabez Christopher, Putrevu Venkata Sai Charan, Rajib Ranjan Maiti

专题命中幻觉与事实性：trustworthy(abstract)

2509.04052 2025-09-05 cs.IR 50%

Safeguarding Patient Trust in the Age of AI: Tackling Health Misinformation with Explainable AI

Sueun Hong, Shuojie Fu, Ovidiu Serban, Brianna Bao, James Kinross, Francesa Toni, Guy Martin, Uddhav Vaghela

专题命中幻觉与事实性：safety(abstract)

2509.03693 2025-09-05 cs.HC cs.MM 50%

Designing Effective AI Explanations for Misinformation Detection: A Comparative Study of Content, Social, and Combined Explanations

Yeaeun Gong, Yifan Liu, Lanyu Shang, Na Wei, Dong Wang

专题命中幻觉与事实性：alignment(abstract)

Comments To appear at CSCW 2025

2507.17187 2025-09-03 cs.GT cs.IR econ.TH 50%

Optimal Calibrated Signaling in Digital Auctions

Zhicheng Du, Wei Tang, Zihe Wang, Shuo Zhang

专题命中幻觉与事实性：trustworthy(abstract)

Comments 56 pages, 5 figures

2506.16988 2025-09-03 cs.IR 50%

RAGentA: Multi-Agent Retrieval-Augmented Generation for Attributed Question Answering

Ines Besrour, Jingbo He, Tobias Schreieder, Michael Färber

专题命中幻觉与事实性：trustworthy(abstract)

Comments Accepted at SIGIR 2025

2501.14179 2025-09-01 cs.HC 50%

AI Chatbots as Professional Service Agents: Developing a Professional Identity

Wenwen Li, Kangwei Shi, Yidong Chai

专题命中幻觉与事实性：alignment(abstract)

2508.16974 2025-08-26 cs.CV 50%

Hierarchical Contextual Grounding LVLM: Enhancing Fine-Grained Visual-Language Understanding with Robust Grounding

Leilei Guo, Antonio Carlos Rivera, Peiyu Tang, Haoxuan Ren, Zheyu Song

机构 * Zhongkai University of Agriculture and Engineering（仲恺农业工程大学）； EDP University of Puerto Rico: San Sebastian（波多黎各圣塞巴斯蒂安EDP大学）

专题命中幻觉与事实性：alignment(abstract)

2506.12609 2025-08-19 cs.CV 50%

Not All Tokens and Heads Are Equally Important: Dual-Level Attention Intervention for Hallucination Mitigation

Lexiang Tang, Xianwei Zhuang, Bang Yang, Zhiyuan Hu, Hongxiang Li, Lu Ma, Jinghan Ru, Yuexian Zou

专题命中幻觉与事实性：alignment(abstract)

2508.11080 2025-08-18 eess.SY cs.SY 50%

Managing Risks from Large Digital Loads Using Coordinated Grid-Forming Storage Network

Soumya Kundu, Kaustav Chatterjee, Ramij R. Hossain, Sai Pushpak Nandanoori, Veronica Adetola

专题命中幻觉与事实性：safety(abstract)

Comments Submitted to IEEE PES T&D Conference and Expo 2026

2503.06776 2025-08-15 cs.RO cs.SY eess.SY 50%

Chance-constrained Linear Quadratic Gaussian Games for Multi-robot Interaction under Uncertainty

Kai Ren, Giulio Salizzoni, Mustafa Emre Gürsoy, Maryam Kamgarpour

机构 * SYCAMORE Lab, École Polytechnique Fédérale de Lausanne (EPFL)（SYCAMORE实验室，瑞士联邦理工学院（EPFL））

专题命中幻觉与事实性：safety(abstract)

Comments Published in IEEE Control Systems Letters

Journal ref IEEE Control Systems Letters, vol. 9, pp. 2061-2066, 2025

2508.03007 2025-08-06 cs.CV 50%

Multi-Granularity Feature Calibration via VFM for Domain Generalized Semantic Segmentation

Xinhui Li, Xiaojie Guo

机构 * Xinhui Li（李新会）； Xiaojie Guo（郭小杰）

专题命中幻觉与事实性：alignment(abstract)

2502.06593 2025-08-05 cs.CV 50%

SAGI: Semantically Aligned and Uncertainty Guided AI Image Inpainting

Paschalis Giakoumoglou, Dimitrios Karageorgiou, Symeon Papadopoulos, Panagiotis C. Petrantonakis

机构 * Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki（阿尔伯塔大学电气与计算机工程系）； Information Technologies Institute, CERTH（信息科技研究所）

专题命中幻觉与事实性：alignment(abstract)

Comments ICCV2025

2412.14369 2025-08-05 eess.SP 50%

Uncertainty Awareness in Wireless Communications and Sensing

Shixiong Wang, Wei Dai, Jianyong Sun, Zongben Xu, Geoffrey Ye Li

专题命中幻觉与事实性：trustworthy(abstract)

Journal ref IEEE Communications Magazine, July 2025

2507.20293 2025-08-01 cs.RO 50%

Decentralized Uncertainty-Aware Multi-Agent Collision Avoidance with Model Predictive Path Integral

Stepan Dergachev, Konstantin Yakovlev

机构 * FRC CSC RAS（俄罗斯科学院应用数学与控制论研究所）； HSE University（俄罗斯高等经济大学）； AIRI（人工智能研究所）

专题命中幻觉与事实性：safety(abstract)

Comments This is a pre-print of the paper accepted to IROS2025. The manuscript includes 8 pages, 4 figures, and 1 table. A supplementary video is available at https://youtu.be/_D4zDYJ4KCk Updated version: added link to source code in the abstract; updated experimental results description in Section VI.A; updated author affiliation and funding information; minor typo corrections

2507.17722 2025-07-24 cs.CV 50%

BetterCheck: Towards Safeguarding VLMs for Automotive Perception Systems

Malsha Ashani Mahawatta Dona, Beatriz Cabrero-Daniel, Yinan Yu, Christian Berger

机构 * University of Gothenburg（哥德堡大学）； Chalmers University of Technology（查尔姆斯理工大学）

专题命中幻觉与事实性：safety(abstract)

Comments Accepted in The IEEE International Conference on Intelligent Transportation Systems (ITSC)2025

2507.09184 2025-07-24 cs.CV 50%

MCA-LLaVA: Manhattan Causal Attention for Reducing Hallucination in Large Vision-Language Models

Qiyan Zhao, Xiaofeng Zhang, Yiheng Li, Yun Xing, Xiaosong Yuan, Feilong Tang, Sinan Fan, Xuhang Chen, Xuyao Zhang, Dahan Wang

机构 * FKLPRIU, Xiamen University of Technology, China（福克斯理工国际大学，厦门理工学院，中国）； Shanghai Jiao Tong University, China（上海交通大学，中国）； Nanyang Technological University, Singapore（南洋理工大学，新加坡）； Jilin University, China（吉林大学，中国）； Monash University, Australia（墨尔本大学，澳大利亚）； Zhejiang University, China（浙江大学，中国）； Huizhou University, China（惠州大学，中国）； Chinese Academy of Sciences, China（中国科学院，中国）

专题命中幻觉与事实性：alignment(abstract)

Comments Accepted in ACM MM 2025

2507.11525 2025-07-16 cs.RO cs.HC 50%

LLM-based ambiguity detection in natural language instructions for collaborative surgical robots

Ana Davila, Jacinto Colan, Yasuhisa Hasegawa

机构 * Institutes of Innovation for Future Society, Nagoya University（未来社会创新研究所，名古屋大学）； Department of Micro-Nano Mechanical Science and Engineering, Nagoya University（微纳米机械科学与工程系，名古屋大学）

专题命中幻觉与事实性：safety(abstract)

Comments Accepted at 2025 IEEE International Conference on Robot and Human Interactive Communication (ROMAN)

Journal ref 2025 IEEE International Conference on Robot and Human Interactive Communication (ROMAN)

2507.09209 2025-07-15 cs.CV 50%

Uncertainty-Driven Expert Control: Enhancing the Reliability of Medical Vision-Language Models

Xiao Liang, Di Wang, Zhicheng Jiao, Ronghan Li, Pengfei Yang, Quan Wang, Tat-Seng Chua

机构 * The Key Laboratory of Smart Human-Computer Interaction and Wearable Technology of Shaanxi Province, Xidian University, China（陕西省智能人机交互与可穿戴技术重点实验室，西安电子科技大学）； Warren Alpert Medical School, Brown University, USA（布朗大学沃伦·阿尔珀特医学院）； National University of Singapore, Singapore（新加坡国立大学）

专题命中幻觉与事实性：alignment(abstract)

2507.07682 2025-07-11 cs.SE 50%

Prompt Engineering for Requirements Engineering: A Literature Review and Roadmap

Kaicheng Huang, Fanyu Wang, Yutan Huang, Chetan Arora

专题命中幻觉与事实性：trustworthy(abstract)

2505.24007 2025-06-30 cs.CV 50%

Preemptive Hallucination Reduction: An Input-Level Approach for Multimodal Language Model

Nokimul Hasan Arif, Shadman Rabby, Md Hefzul Hossain Papon, Sabbir Ahmed

专题命中幻觉与事实性：trustworthy(abstract)

Comments Submitted for review in NCAA Springer, 21 pages, 4 figures, 4 Tables

2506.17901 2025-06-24 cs.CV 50%

PostAlign: Multimodal Grounding as a Corrective Lens for MLLMs

Yixuan Wu, Yang Zhang, Jian Wu, Philip Torr, Jindong Gu

机构 * University of Oxford（牛津大学）； Zhejiang University（浙江大学）； National University of Singapore（新加坡国立大学）

专题命中幻觉与事实性：alignment(abstract)

2506.17664 2025-06-24 cs.CV 50%

MDSAM:Memory-Driven Sparse Attention Matrix for LVLMs Hallucination Mitigation

Shuaiye Lu, Linjiang Zhou, Xiaochuan Shi

机构 * Wuhan University（武汉大学）

专题命中幻觉与事实性：alignment(abstract)

2506.13038 2025-06-18 cs.CV cs.MM 50%

HKD4VLM: A Progressive Hybrid Knowledge Distillation Framework for Robust Multimodal Hallucination and Factuality Detection in VLMs

Zijian Zhang, Xuecheng Wu, Danlei Huang, Siyu Yan, Chong Peng, Xuezhi Cao

机构 * Xi'an Jiaotong University（西安交通大学）； East China Normal University（华东师范大学）

专题命中幻觉与事实性：alignment(abstract)

2506.12507 2025-06-17 cs.RO 50%

Sense and Sensibility: What makes a social robot convincing to high-school students?

Pablo Gonzalez-Oliveras, Olov Engwall, Ali Reza Majlesi

机构 * Dept. of Intelligent Systems, KTH Royal Institute of Technology, Stockholm, Sweden（智能系统系，皇家理工学院）； Dept. of Education, Stockholm University, Stockholm, Sweden（教育系，斯德哥尔摩大学）

专题命中幻觉与事实性：alignment(abstract)

Comments 14 pages; 8 figures; 3 tables; RSS 2025 (Robotics: Science & Systems)

2411.08329 2025-06-12 eess.SY cs.SY 50%

Neural Network Certification Informed Power System Transient Stability Preventive Control with Renewable Energy

Tong Su, Junbo Zhao

专题命中幻觉与事实性：safety(abstract)

Journal ref IEEE Transactions on Power Systems, 2025

1. 幻觉与事实性 1756 篇

No Concept Left Behind: Test-Time Optimization for Compositional Text-to-Image Generation

Situational Awareness for Safe and Robust Multi-Agent Interactions Under Uncertainty

Co-STAR: Collaborative Curriculum Self-Training with Adaptive Regularization for Source-Free Video Domain Adaptation

AD-DINOv3: Enhancing DINOv3 for Zero-Shot Anomaly Detection with Anomaly-Aware Calibration

Occupancy-aware Trajectory Planning for Autonomous Valet Parking in Uncertain Dynamic Environments

rCamInspector: Building Reliability and Trust on IoT (Spy) Camera Detection using XAI

Safeguarding Patient Trust in the Age of AI: Tackling Health Misinformation with Explainable AI

Designing Effective AI Explanations for Misinformation Detection: A Comparative Study of Content, Social, and Combined Explanations

Optimal Calibrated Signaling in Digital Auctions

RAGentA: Multi-Agent Retrieval-Augmented Generation for Attributed Question Answering

AI Chatbots as Professional Service Agents: Developing a Professional Identity

Hierarchical Contextual Grounding LVLM: Enhancing Fine-Grained Visual-Language Understanding with Robust Grounding

Not All Tokens and Heads Are Equally Important: Dual-Level Attention Intervention for Hallucination Mitigation

Managing Risks from Large Digital Loads Using Coordinated Grid-Forming Storage Network

Chance-constrained Linear Quadratic Gaussian Games for Multi-robot Interaction under Uncertainty

Multi-Granularity Feature Calibration via VFM for Domain Generalized Semantic Segmentation

SAGI: Semantically Aligned and Uncertainty Guided AI Image Inpainting

Uncertainty Awareness in Wireless Communications and Sensing

Decentralized Uncertainty-Aware Multi-Agent Collision Avoidance with Model Predictive Path Integral

BetterCheck: Towards Safeguarding VLMs for Automotive Perception Systems

MCA-LLaVA: Manhattan Causal Attention for Reducing Hallucination in Large Vision-Language Models

LLM-based ambiguity detection in natural language instructions for collaborative surgical robots

Uncertainty-Driven Expert Control: Enhancing the Reliability of Medical Vision-Language Models

Prompt Engineering for Requirements Engineering: A Literature Review and Roadmap

Preemptive Hallucination Reduction: An Input-Level Approach for Multimodal Language Model

PostAlign: Multimodal Grounding as a Corrective Lens for MLLMs

MDSAM:Memory-Driven Sparse Attention Matrix for LVLMs Hallucination Mitigation

HKD4VLM: A Progressive Hybrid Knowledge Distillation Framework for Robust Multimodal Hallucination and Factuality Detection in VLMs

Sense and Sensibility: What makes a social robot convincing to high-school students?

Neural Network Certification Informed Power System Transient Stability Preventive Control with Renewable Energy