视觉大模型 / VLM - arXivDaily 专题

2506.01118 2025-06-03 cs.CV 57%

Revolutionizing Radiology Workflow with Factual and Efficient CXR Report Generation

Pimchanok Sukjai, Apiradee Boonmee

机构 * Kasem Bundit University（Kasem Bundit大学）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

2505.24693 2025-06-02 cs.CV 57%

Conformal Prediction for Zero-Shot Models

Julio Silva-Rodríguez, Ismail Ben Ayed, Jose Dolz

机构 * ÉTS Montréal（蒙特利尔ÉTS）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

Comments CVPR 2025. Code: https://github.com/jusiro/CLIP-Conformal

2505.24519 2025-06-02 cs.CV cs.CL 57%

AMIA: Automatic Masking and Joint Intention Analysis Makes LVLMs Robust Jailbreak Defenders

Yuqi Zhang, Yuchun Miao, Zuchao Li, Liang Ding

机构 * School of Computer Science, Wuhan University（武汉大学计算机学院）； School of Artificial Intelligence, Wuhan University（武汉大学人工智能学院）； The University of Sydney（悉尼大学）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

Comments 11 pages, 7 figures

2505.24458 2025-06-02 cs.AI 57%

SEAR: A Multimodal Dataset for Analyzing AR-LLM-Driven Social Engineering Behaviors

Tianlong Yu, Chenghang Ye, Zheyu Yang, Ziyi Zhou, Cui Tang, Zui Tao, Jun Zhang, Kailong Wang, Liting Zhou, Yang Yang, Ting Bi

机构 * Huazhong University of Science and Technology（华中科技大学）； Hubei University（湖北大学）； Dublin City University（都柏林城市大学）

专题命中幻觉与鲁棒性：multimodal large language model(abstract)；分类 cs.AI

2505.24317 2025-06-02 cs.LG 57%

ROAD: Responsibility-Oriented Reward Design for Reinforcement Learning in Autonomous Driving

Yongming Chen, Miner Chen, Liewen Liao, Mingyang Jiang, Xiang Zuo, Hengrui Zhang, Yuchen Xi, Songan Zhang

机构 * Global Institute of Future Technology（未来技术全球研究院）； Shanghai Jiao Tong University（上海交通大学）； College of Humanities（人文学院）； Department of Automation（自动化系）； School of Aeronautics and Astronautics（航空宇航学院）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.LG

2410.10112 2025-05-30 cs.CV cs.CL 57%

Can We Predict Performance of Large Models across Vision-Language Tasks?

Qinyu Zhao, Ming Xu, Kartik Gupta, Akshay Asthana, Liang Zheng, Stephen Gould

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

Comments ICML2025. Project page: https://github.com/Qinyu-Allen-Zhao/CrossPred-LVLM

2505.21106 2025-05-28 cs.AI 57%

Interpreting Social Bias in LVLMs via Information Flow Analysis and Multi-Round Dialogue Evaluation

Zhengyang Ji, Yifan Jia, Shang Gao, Yutao Yue

机构 * The Hong Kong University of Science and Technology (Guangzhou)（香港科学与技术大学（广州））； Shandong University（山东大学）； Institute of Deep Perception Technology, JITRI（深度感知技术研究所）

专题命中幻觉与鲁棒性：vision language model(abstract)；分类 cs.AI

2505.18530 2025-05-27 cs.MA cs.AI 57%

MRGAgents: A Multi-Agent Framework for Improved Medical Report Generation with Med-LVLMs

Pengyu Wang, Shuchang Ye, Usman Naseem, Jinman Kim

机构 * University of Sydney（悉尼大学）； Macquarie University（麦考瑞大学）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.AI

Comments 10pages

2505.17727 2025-05-26 cs.CV 57%

SafeMVDrive: Multi-view Safety-Critical Driving Video Synthesis in the Real World Domain

Jiawei Zhou, Linye Lyu, Zhuotao Tian, Cheng Zhuo, Yu Li

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

2505.16778 2025-05-23 cs.CV 57%

Single Domain Generalization for Few-Shot Counting via Universal Representation Matching

Xianing Chen, Si Huo, Borui Jiang, Hailin Hu, Xinghao Chen

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

Comments CVPR 2025

2505.16193 2025-05-23 cs.CL cs.CV 57%

An Empirical Study on Configuring In-Context Learning Demonstrations for Unleashing MLLMs' Sentimental Perception Capability

Daiqing Wu, Dongbao Yang, Sicheng Zhao, Can Ma, Yu Zhou

专题命中幻觉与鲁棒性：multimodal large language model(abstract)；分类 cs.CV

2408.10202 2025-05-22 cs.CV 57%

SANER: Annotation-free Societal Attribute Neutralizer for Debiasing CLIP

Yusuke Hirota, Min-Hung Chen, Chien-Yi Wang, Yuta Nakashima, Yu-Chiang Frank Wang, Ryo Hachiuma

机构 * NVIDIA ； Osaka University（大阪大学）； National Taiwan University（国立台湾大学）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

Comments ICLR 2025

2502.05863 2025-05-21 cs.IR cs.AI cs.MM 57%

Uni-Retrieval: A Multi-Style Retrieval Framework for STEM's Education

Yanhao Jia, Xinyi Wu, Hao Li, Qinglin Zhang, Yuxiao Hu, Shuai Zhao, Wenqi Fan

机构 * Nanyang Technological University, Singapore（南洋理工大学）； Shanghai Jiao Tong University, China（上海交通大学）； Peking University, China（北京大学）； Hong Kong Polytechnic University, Hong Kong, China（香港理工大学）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.AI

2505.12826 2025-05-20 cs.CV 57%

Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering

Jianfeng Cai, Wengang Zhou, Zongmeng Zhang, Jiale Hong, Nianji Zhan, Houqiang Li

机构 * University of Science and Technology of China（中国科学技术大学）； Shanghai Jiaotong University（上海交通大学）； Merchants Union Consumer Finance Company Limited

专题命中幻觉与鲁棒性：multimodal large language model(abstract)；分类 cs.CV

2410.23687 2025-05-20 cs.CV cs.CR 57%

Adversarial Attacks of Vision Tasks in the Past 10 Years: A Survey

Chiyu Zhang, Lu Zhou, Xiaogang Xu, Jiafei Wu, Zhe Liu

机构 * Nanjing University of Aeronautics and Astronautics（南京航空航天大学）； The Chinese University of Hong Kong（香港中文大学）； The University of Hong Kong（香港大学）； Zhejiang Lab, Nanjing University of Aeronautics and Astronautics（浙江实验室，南京航空航天大学）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

2505.07172 2025-05-13 cs.CV 57%

Critique Before Thinking: Mitigating Hallucination through Rationale-Augmented Instruction Tuning

Zexian Yang, Dian Li, Dayan Wu, Gang Liu, Weiping Wang

机构 * Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）； Foundation Technology Center, Tencent PCG（腾讯PCG基础技术中心）

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

2505.06079 2025-05-12 cs.RO cs.CV 57%

TREND: Tri-teaching for Robust Preference-based Reinforcement Learning with Demonstrations

Shuaiyi Huang, Mara Levy, Anubhav Gupta, Daniel Ekpo, Ruijie Zheng, Abhinav Shrivastava

机构 * University of Maryland, College Park（马里兰大学学院公园分校）

专题命中幻觉与鲁棒性：VLM(abstract)；分类 cs.CV

Comments ICRA 2025

2503.20698 2025-05-12 cs.CV cs.IR 57%

MMMORRF: Multimodal Multilingual Modularized Reciprocal Rank Fusion

Saron Samuel, Dan DeGenaro, Jimena Guallar-Blasco, Kate Sanders, Oluwaseun Eisape, Tanner Spendlove, Arun Reddy, Alexander Martin, Andrew Yates, Eugene Yang, Cameron Carpenter, David Etter, Efsun Kayi, Matthew Wiesner, Kenton Murray, Reno Kriz

机构 * Stanford University（斯坦福大学）； Georgetown University（乔治城大学）； Johns Hopkins University（约翰霍普金斯大学）； UC Berkeley（伯克利大学）； BYU

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

2504.15619 2025-04-23 cs.CV 57%

AdaViP: Aligning Multi-modal LLMs via Adaptive Vision-enhanced Preference Optimization

Jinda Lu, Jinghan Li, Yuan Gao, Junkang Wu, Jiancan Wu, Xiang Wang, Xiangnan He

机构 * University of Science and Technology of China（中国科学技术大学）

专题命中幻觉与鲁棒性：multimodal large language model(abstract)；分类 cs.CV

2504.14526 2025-04-22 cs.CV cs.CL 57%

Are Vision LLMs Road-Ready? A Comprehensive Benchmark for Safety-Critical Driving Video Understanding

Tong Zeng, Longfeng Wu, Liang Shi, Dawei Zhou, Feng Guo

机构 * Department of Computer Science, Virginia Tech（弗吉尼亚理工学院计算机科学系）； Virginia Tech Transportation Institute, Virginia Tech（弗吉尼亚理工学院交通研究所）； Department of Statistics, Virginia Tech（弗吉尼亚理工学院统计学系）

专题命中幻觉与鲁棒性：visual question answering(abstract)；分类 cs.CV

2504.01081 2025-04-10 cs.CV cs.CL eess.IV 57%

ShieldGemma 2: Robust and Tractable Image Content Moderation

Wenjun Zeng, Dana Kurniawan, Ryan Mullins, Yuchi Liu, Tamoghna Saha, Dirichi Ike-Njoku, Jindong Gu, Yiwen Song, Cai Xu, Jingjing Zhou, Aparna Joshi, Shravan Dheep, Mani Malek, Hamid Palangi, Joon Baek, Rick Pereira, Karthik Narasimhan

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

2504.05324 2025-04-09 cs.IR cs.AI cs.CL 57%

Hybrid Retrieval for Hallucination Mitigation in Large Language Models: A Comparative Analysis

Chandana Sree Mala, Gizem Gezici, Fosca Giannotti

专题命中幻觉与鲁棒性：grounding(abstract)；分类 cs.AI

2504.04225 2025-04-08 cs.CV 57%

Resilience of Vision Transformers for Domain Generalisation in the Presence of Out-of-Distribution Noisy Images

Hamza Riaz, Alan F. Smeaton

专题命中幻觉与鲁棒性：grounding(abstract)；分类 cs.CV

Comments 31 pages

2504.00429 2025-04-02 cs.CV 57%

Unleashing the Power of Pre-trained Encoders for Universal Adversarial Attack Detection

Yinghe Zhang, Chi Liu, Shuai Zhou, Sheng Shen, Peng Gui

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

2503.20673 2025-03-28 cs.CV 57%

Mitigating Low-Level Visual Hallucinations Requires Self-Awareness: Database, Model and Training Strategy

Yinan Sun, Xiongkuo Min, Zicheng Zhang, Yixuan Gao, Yuqin Cao, Guangtao Zhai

专题命中幻觉与鲁棒性：multimodal large language model(abstract)；分类 cs.CV

2411.04713 2025-03-21 cs.CV 57%

Multi-Reward as Condition for Instruction-based Image Editing

Xin Gu, Ming Li, Libo Zhang, Fan Chen, Longyin Wen, Tiejian Luo, Sijie Zhu

专题命中幻觉与鲁棒性：vision language model(abstract)；分类 cs.CV

2503.15092 2025-03-20 cs.CR cs.AI cs.CL 57%

Towards Understanding the Safety Boundaries of DeepSeek Models: Evaluation and Findings

Zonghao Ying, Guangyi Zheng, Yongxin Huang, Deyue Zhang, Wenxin Zhang, Quanchen Zou, Aishan Liu, Xianglong Liu, Dacheng Tao

专题命中幻觉与鲁棒性：multimodal large language model(abstract)；分类 cs.AI

2406.00430 2025-03-18 cs.RO cs.AI 57%

Evaluating Uncertainty-based Failure Detection for Closed-Loop LLM Planners

Zhi Zheng, Qian Feng, Hang Li, Alois Knoll, Jianxiang Feng

专题命中幻觉与鲁棒性：multimodal large language model(abstract)；分类 cs.AI

Comments Accepted at ICRA 2024 Workshop on Back to the Future: Robot Learning Going Probabilistic. Website: https://sites.google.com/view/konwloop/home

2503.08216 2025-03-17 cs.CV 57%

Attention Hijackers: Detect and Disentangle Attention Hijacking in LVLMs for Hallucination Mitigation

Beitao Chen, Xinyu Lyu, Lianli Gao, Jingkuan Song, Heng Tao Shen

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.CV

2409.15658 2025-03-14 cs.RO cs.AI 57%

Long-horizon Embodied Planning with Implicit Logical Inference and Hallucination Mitigation

Siyuan Liu, Jiawei Du, Sicheng Xiang, Zibo Wang, Dingsheng Luo

专题命中幻觉与鲁棒性：vision-language model(abstract)；分类 cs.AI

1. 幻觉与鲁棒性 2277 篇

Revolutionizing Radiology Workflow with Factual and Efficient CXR Report Generation

Conformal Prediction for Zero-Shot Models

AMIA: Automatic Masking and Joint Intention Analysis Makes LVLMs Robust Jailbreak Defenders

SEAR: A Multimodal Dataset for Analyzing AR-LLM-Driven Social Engineering Behaviors

ROAD: Responsibility-Oriented Reward Design for Reinforcement Learning in Autonomous Driving

Can We Predict Performance of Large Models across Vision-Language Tasks?

Interpreting Social Bias in LVLMs via Information Flow Analysis and Multi-Round Dialogue Evaluation

MRGAgents: A Multi-Agent Framework for Improved Medical Report Generation with Med-LVLMs

SafeMVDrive: Multi-view Safety-Critical Driving Video Synthesis in the Real World Domain

Single Domain Generalization for Few-Shot Counting via Universal Representation Matching

An Empirical Study on Configuring In-Context Learning Demonstrations for Unleashing MLLMs' Sentimental Perception Capability

SANER: Annotation-free Societal Attribute Neutralizer for Debiasing CLIP

Uni-Retrieval: A Multi-Style Retrieval Framework for STEM's Education

Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering

Adversarial Attacks of Vision Tasks in the Past 10 Years: A Survey

Critique Before Thinking: Mitigating Hallucination through Rationale-Augmented Instruction Tuning

TREND: Tri-teaching for Robust Preference-based Reinforcement Learning with Demonstrations

MMMORRF: Multimodal Multilingual Modularized Reciprocal Rank Fusion

AdaViP: Aligning Multi-modal LLMs via Adaptive Vision-enhanced Preference Optimization

Are Vision LLMs Road-Ready? A Comprehensive Benchmark for Safety-Critical Driving Video Understanding

ShieldGemma 2: Robust and Tractable Image Content Moderation

Hybrid Retrieval for Hallucination Mitigation in Large Language Models: A Comparative Analysis

Resilience of Vision Transformers for Domain Generalisation in the Presence of Out-of-Distribution Noisy Images

Unleashing the Power of Pre-trained Encoders for Universal Adversarial Attack Detection

Mitigating Low-Level Visual Hallucinations Requires Self-Awareness: Database, Model and Training Strategy

Multi-Reward as Condition for Instruction-based Image Editing

Towards Understanding the Safety Boundaries of DeepSeek Models: Evaluation and Findings

Evaluating Uncertainty-based Failure Detection for Closed-Loop LLM Planners

Attention Hijackers: Detect and Disentangle Attention Hijacking in LVLMs for Hallucination Mitigation

Long-horizon Embodied Planning with Implicit Logical Inference and Hallucination Mitigation