多模态大模型 - arXivDaily 专题

2511.05057 2025-11-10 cs.CV 70%

Role-SynthCLIP: A Role Play Driven Diverse Synthetic Data Approach

Yuanxiang Huangfu, Chaochao Wang, Weilei Wang

机构 * PatSnap Co., LTD.（PatSnap公司）

专题命中图文多模态：multimodal(abstract)；image-text(abstract)；分类 cs.CV

2510.21849 2025-11-07 cs.LG cs.AI 70%

TowerVision: Understanding and Improving Multilinguality in Vision-Language Models

André G. Viveiros, Patrick Fernandes, Saul Santos, Sonal Sannigrahi, Emmanouil Zaranis, Nuno M. Guerreiro, Amin Farajian, Pierre Colombo, Graham Neubig, André F. T. Martins

机构 * Instituto Superior Técnico, Universidade de Lisboa（里斯本大学技术高级学院）； Instituto de Telecomunicações（电信研究所）； Carnegie Mellon University（卡内基梅隆大学）； Sword Health（Sword健康）； TransPerfect ； MICS, CentraleSupélec, Université Paris-Saclay（MICS，中央圣艾尔布里大学，巴黎萨克雷大学）； ELLIS Unit Lisbon（里斯本ELLIS单位）

专题命中图文多模态：multimodal(abstract)；image-text(abstract)；分类 cs.AI

Comments 15 pages, 7 figures, submitted to arXiv October 2025. All models, datasets, and training code will be released at https://huggingface.co/collections/utter-project/towervision

2506.04220 2025-11-06 cs.CV 70%

Struct2D: A Perception-Guided Framework for Spatial Reasoning in MLLMs

Fangrui Zhu, Hanhui Wang, Yiming Xie, Jing Gu, Tianye Ding, Jianwei Yang, Huaizu Jiang

机构 * Northeastern University（东北大学）； Microsoft Research（微软研究院）； University of Southern California（南加州大学）； University of California, Santa Cruz（加州大学圣克鲁兹分校）

专题命中图文多模态：multimodal(abstract)；MLLM(abstract)；分类 cs.CV

Comments NeurIPS 2025, code link: https://github.com/neu-vi/struct2d

2510.25070 2025-10-30 cs.CV 70%

Vision-Language Integration for Zero-Shot Scene Understanding in Real-World Environments

Manjunath Prasad Holenarasipura Rajiv, B. M. Vidyavathi

专题命中图文多模态：multimodal(abstract)；cross-modal(abstract)；分类 cs.CV

Comments Preprint under review at IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025

2510.22785 2025-10-28 cs.CV 70%

Self-Calibrated Consistency can Fight Back for Adversarial Robustness in Vision-Language Models

Jiaxiang Liu, Jiawei Du, Xiao Liu, Prayag Tiwari, Mingkun Xu

机构 * Guangdong Institute of Intelligence Science and Technology（广东智能科学与技术研究院）； Agency for Science, Technology and Research（科技研究局）； School of Information Technology（信息技术学院）

专题命中图文多模态：cross-modal(abstract)；image-text(abstract)；分类 cs.CV

2505.07019 2025-10-28 cs.CV 70%

A Vision-Language Foundation Model for Leaf Disease Identification

Khang Nguyen Quoc, Lan Le Thi Thu, Luyl-Da Quach

专题命中图文多模态：multi-modal(abstract)；image-text(abstract)；分类 cs.CV

2510.21606 2025-10-27 cs.CV 70%

Modest-Align: Data-Efficient Alignment for Vision-Language Models

Jiaxiang Liu, Yuan Wang, Jiawei Du, Joey Tianyi Zhou, Mingkun Xu, Zuozhu Liu

机构 * Guangdong Institute of Intelligence Science and Technology（广东智能科学与技术研究院）； ZJU-Angelalign R&D Center for Intelligence Healthcare（浙大天使align智能医疗研发中心）； Centre for Frontier AI Research (CFAR)（前沿人工智能研究中心）； Agency for Science, Technology and Research (A*STAR)（科技研究局）； Institute of High Performance Computing (IHPC)（高性能计算研究所）

专题命中图文多模态：cross-modal(abstract)；image-text(abstract)；分类 cs.CV

2510.21311 2025-10-27 cs.CV 70%

FineRS: Fine-grained Reasoning and Segmentation of Small Objects with Reinforcement Learning

Lu Zhang, Jiazuo Yu, Haomiao Xiong, Ping Hu, Yunzhi Zhuge, Huchuan Lu, You He

机构 * Dalian University of Technology（大连理工大学）； University of Electronic Science and Technology of China（电子科技大学）； Tsinghua Shenzhen International Graduate School（清华大学深圳国际graduate school）

专题命中图文多模态：multi-modal(abstract)；MLLM(abstract)；分类 cs.CV

Comments Accepted to NeurIPS 2025

2510.16870 2025-10-21 cs.CV 70%

Uncovering Brain-Like Hierarchical Patterns in Vision-Language Models through fMRI-Based Neural Encoding

Yudan Ren, Xinlong Wang, Kexin Wang, Tian Xia, Zihan Ma, Zhaowei Li, Xiangrong Bi, Xiao Li, Xiaowei He

专题命中图文多模态：multimodal(abstract)；cross-modal(abstract)；分类 cs.CV

Comments 14 pages, 7 figures

2505.13946 2025-10-21 cs.AI 70%

Visual Instruction Bottleneck Tuning

Changdae Oh, Jiatong Li, Shawn Im, Sharon Li

机构 * Department of Computer Sciences, University of Wisconsin–Madison（计算机科学系，威斯康星大学麦迪逊分校）

专题命中图文多模态：multimodal(abstract)；MLLM(abstract)；分类 cs.AI

Comments NeurIPS 2025

2510.13235 2025-10-16 cs.CV 70%

EPIPTrack: Rethinking Prompt Modeling with Explicit and Implicit Prompts for Multi-Object Tracking

Yukuan Zhang, Jiarui Zhao, Shangqing Nie, Jin Kuang, Shengsheng Wang

机构 * College of Computer Science and Technology, Jilin University（吉林大学计算机科学与技术学院）； Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University（吉林大学教育部长春符号计算与知识工程重点实验室）； Yangtze University（扬子大学）

专题命中图文多模态：multimodal(abstract)；cross-modal(abstract)；分类 cs.CV

2507.01738 2025-10-14 cs.CV 70%

DeRIS: Decoupling Perception and Cognition for Enhanced Referring Image Segmentation through Loopback Synergy

Ming Dai, Wenxuan Cheng, Jiang-jiang Liu, Sen Yang, Wenxiao Cai, Yanpeng Sun, Wankou Yang

机构 * Southeast University（东南大学）； Baidu VIS（百度VIS）； Stanford University（斯坦福大学）

专题命中图文多模态：multi-modal(abstract)；image-text(abstract)；分类 cs.CV

Comments Accepted by ICCV 2025

2502.15130 2025-10-10 cs.CV 70%

TransMamba: Fast Universal Architecture Adaption from Transformers to Mamba

Xiuwei Chen, Wentao Hu, Xiao Dong, Sihao Lin, Zisheng Chen, Meng Cao, Yina Zhuang, Jianhua Han, Hang Xu, Xiaodan Liang

机构 * Shenzhen Campus of Sun Yat-sen University（中山大学深圳校区）； Hong Kong Polytechnic University（香港理工大学）； Zhuhai Campus of Sun Yat-sen University（中山大学珠海校区）； University of Adelaide（阿德莱德大学）； Peking University（北京大学）； Huawei Noah’s Ark Lab（华为诺亚实验室）

专题命中图文多模态：multimodal(abstract)；multi-modal(abstract)；分类 cs.CV

2510.02815 2025-10-06 cs.CV 70%

Med-K2N: Flexible K-to-N Modality Translation for Medical Image Synthesis

Feng Yuan, Yifan Gao, Yuehua Ye, Haoyue Li, Xin Gao

机构 * University of Science and Technology of China（中国科学技术大学）； Suzhou Institute of Biomedical Engineering and Technology（苏州生物医学工程与技术研究所）； Chinese Academy of Sciences（中国科学院）； The Third Affiliated Hospital of Sun Yat-sen University（中山大学第三附属医院）

专题命中图文多模态：multi-modal(abstract)；cross-modal(abstract)；分类 cs.CV

Comments ICLR2026 under review

2509.25654 2025-10-01 cs.CV 70%

DescribeEarth: Describe Anything for Remote Sensing Images

Kaiyu Li, Zixuan Jiang, Xiangyong Cao, Jiayu Wang, Yuchen Xiao, Deyu Meng, Zhi Wang

机构 * School of Software Engineering, Xi’an Jiaotong University（西安交通大学软件工程学院）； College of Artificial Intelligence, Xi’an Jiaotong University（西安交通大学人工智能学院）； School of Computer Science and Technology and Ministry of Education Key Lab For Intelligent Networks and Network Security, Xi’an Jiaotong University（西安交通大学计算机科学与技术学院和教育部智能网络与网络安全重点实验室）； School of Mathematics and Statistics and Ministry of Education Key Lab of Intelligent Networks and Network Security, Xi’an Jiaotong University（西安交通大学数学与统计学院和教育部智能网络与网络安全重点实验室）； Pazhou Laboratory (Huangpu), Guangzhou, Guangdong, China（琶洲实验室（黄埔），广州，广东，中国）

专题命中图文多模态：multi-modal(abstract)；MLLM(abstract)；分类 cs.CV

2509.06105 2025-10-01 cs.CV 70%

PathoHR: Hierarchical Reasoning for Vision-Language Models in Pathology

Yating Huang, Ziyan Huang, Lintao Xiang, Qijun Yang, Hujun Yin

机构 * University of Manchester（曼彻斯特大学）； South China University of Technology（华南理工大学）

专题命中图文多模态：multimodal(abstract)；cross-modal(abstract)；分类 cs.CV

Comments Accept by EMNLP2025

2509.24258 2025-09-30 cs.CV 70%

When MLLMs Meet Compression Distortion: A Coding Paradigm Tailored to MLLMs

Jinming Liu, Zhaoyang Jia, Jiahao Li, Bin Li, Xin Jin, Wenjun Zeng, Yan Lu

机构 * Shanghai Jiao Tong University（上海交通大学）； Eastern Institute of Technology（东部技术研究所）； Microsoft Research Asia（微软亚洲研究院）

专题命中图文多模态：multimodal(abstract)；MLLM(abstract)；分类 cs.CV

2509.20119 2025-09-25 cs.CV 70%

A Simple Data Augmentation Strategy for Text-in-Image Scientific VQA

Belal Shoer, Yova Kementchedjhieva

机构 * MBZUAI（穆罕默德·本·拉希德智能研究院）

专题命中图文多模态：multimodal(abstract)；image-text(abstract)；分类 cs.CV

Comments Accepted at WiNLP, 2025

2509.16673 2025-09-23 cs.CV 70%

MedCutMix: A Data-Centric Approach to Improve Radiology Vision-Language Pre-training with Disease Awareness

Sinuo Wang, Yutong Xie, Yuyuan Liu, Qi Wu

机构 * The University of Adelaide（阿德莱德大学）； Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）； University of Oxford（牛津大学）

专题命中图文多模态：multi-modal(abstract)；image-text(abstract)；分类 cs.CV

2509.16560 2025-09-23 cs.CV 70%

Captioning for Text-Video Retrieval via Dual-Group Direct Preference Optimization

Ji Soo Lee, Byungoh Ko, Jaewon Cho, Howoong Lee, Jaewoon Byun, Hyunwoo J. Kim

机构 * Korea University（韩国大学）； Hanwha Vision（翰威英航）； KAIST（韩国科学技术院）

专题命中图文多模态：multi-modal(abstract)；MLLM(abstract)；分类 cs.CV

Comments EMNLP 2025 Findings

2509.15226 2025-09-19 cs.CV 70%

Calibration-Aware Prompt Learning for Medical Vision-Language Models

Abhishek Basu, Fahad Shamshad, Ashshak Sharifdeen, Karthik Nandakumar, Muhammad Haris Khan

机构 * Department of Computer Vision, Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)（计算机视觉系，Mohamed bin Zayed人工智能大学）； Department of Computer Science and Engineering, Michigan State University (MSU)（计算机科学与工程系，密歇根州立大学）

专题命中图文多模态：multimodal(abstract)；image-text(abstract)；分类 cs.CV

Comments Accepted in BMVC 2025

2509.14738 2025-09-19 cs.CL 70%

UnifiedVisual: A Framework for Constructing Unified Vision-Language Datasets

Pengyu Wang, Shaojun Zhou, Chenkun Tan, Xinghao Wang, Wei Huang, Zhen Ye, Zhaowei Li, Botian Jiang, Dong Zhang, Xipeng Qiu

机构 * Fudan University（复旦大学）

专题命中图文多模态：multimodal(abstract)；cross-modal(abstract)；分类 cs.CL

Comments Accepted by Findings of EMNLP2025

2405.16915 2025-09-16 cs.CV cs.LG 70%

Multilingual Diversity Improves Vision-Language Representations

Thao Nguyen, Matthew Wallingford, Sebastin Santy, Wei-Chiu Ma, Sewoong Oh, Ludwig Schmidt, Pang Wei Koh, Ranjay Krishna

机构 * University of Washington（华盛顿大学）； Allen Institute for Artificial Intelligence（人工智能研究院）

专题命中图文多模态：multimodal(abstract)；image-text(abstract)；分类 cs.CV

Comments NeurIPS 2024 Spotlight paper

2509.09397 2025-09-12 cs.CV 70%

Decoupling Clinical and Class-Agnostic Features for Reliable Few-Shot Adaptation under Shift

Umaima Rahman, Raza Imam, Mohammad Yaqub, Dwarikanath Mahapatra

机构 * Mohamed Bin Zayed University of Artificial Intelligence（莫扎德人工智能大学）； Khalifa University（卡利法大学）

专题命中图文多模态：cross-modal(abstract)；image-text(abstract)；分类 cs.CV

2509.08570 2025-09-11 cs.CV 70%

Vision-Language Semantic Aggregation Leveraging Foundation Model for Generalizable Medical Image Segmentation

Wenjun Yu, Yinchen Zhou, Jia-Xuan Jiang, Shubin Zeng, Yuee Li, Zhong Wang

机构 * organization= School of Information Science \& Engineering, Lanzhou University , addressline= , city= Lanzhou , postcode= 730000 , country= China

专题命中图文多模态：multimodal(abstract)；cross-modal(abstract)；分类 cs.CV

Comments 29 pages and 8 figures

2509.04378 2025-09-10 cs.CV 70%

Aesthetic Image Captioning with Saliency Enhanced MLLMs

Yilin Tao, Jiashui Huang, Huaze Xu, Ling Shao

专题命中图文多模态：multimodal(abstract)；MLLM(abstract)；分类 cs.CV

2509.03032 2025-09-04 cs.CV 70%

Background Matters Too: A Language-Enhanced Adversarial Framework for Person Re-Identification

Kaicong Huang, Talha Azfar, Jack M. Reilly, Thomas Guggisberg, Ruimin Ke

专题命中图文多模态：multimodal(abstract)；cross-modal(abstract)；分类 cs.CV

2508.20243 2025-08-29 cs.CV cs.LG 70%

Linking heterogeneous microstructure informatics with expert characterization knowledge through customized and hybrid vision-language representations for industrial qualification

Mutahar Safdar, Gentry Wood, Max Zimmermann, Guy Lamouche, Priti Wanjara, Yaoyao Fiona Zhao

专题命中图文多模态：multi-modal(abstract)；cross-modal(abstract)；分类 cs.CV

Comments 46 pages, 33 figures, Submitted to Advanced Engineering Informatics, under revision

2508.18634 2025-08-28 cs.CV 70%

OwlCap: Harmonizing Motion-Detail for Video Captioning via HMD-270K and Caption Set Equivalence Reward

Chunlin Zhong, Qiuxia Hou, Zhangjun Zhou, Shuang Hao, Haonan Lu, Yanhao Zhang, He Tang, Xiang Bai

专题命中图文多模态：multi-modal(abstract)；MLLM(abstract)；分类 cs.CV

Comments 9 pages, 6figures

2508.11218 2025-08-18 cs.CV cs.LG 70%

A CLIP-based Uncertainty Modal Modeling (UMM) Framework for Pedestrian Re-Identification in Autonomous Driving

Jialin Li, Shuqi Wu, Ning Wang

专题命中图文多模态：multimodal(abstract)；cross-modal(abstract)；分类 cs.CV

1. 图文多模态 4651 篇

Role-SynthCLIP: A Role Play Driven Diverse Synthetic Data Approach

TowerVision: Understanding and Improving Multilinguality in Vision-Language Models

Struct2D: A Perception-Guided Framework for Spatial Reasoning in MLLMs

Vision-Language Integration for Zero-Shot Scene Understanding in Real-World Environments

Self-Calibrated Consistency can Fight Back for Adversarial Robustness in Vision-Language Models

A Vision-Language Foundation Model for Leaf Disease Identification

Modest-Align: Data-Efficient Alignment for Vision-Language Models

FineRS: Fine-grained Reasoning and Segmentation of Small Objects with Reinforcement Learning

Uncovering Brain-Like Hierarchical Patterns in Vision-Language Models through fMRI-Based Neural Encoding

Visual Instruction Bottleneck Tuning

EPIPTrack: Rethinking Prompt Modeling with Explicit and Implicit Prompts for Multi-Object Tracking

DeRIS: Decoupling Perception and Cognition for Enhanced Referring Image Segmentation through Loopback Synergy

TransMamba: Fast Universal Architecture Adaption from Transformers to Mamba

Med-K2N: Flexible K-to-N Modality Translation for Medical Image Synthesis

DescribeEarth: Describe Anything for Remote Sensing Images

PathoHR: Hierarchical Reasoning for Vision-Language Models in Pathology

When MLLMs Meet Compression Distortion: A Coding Paradigm Tailored to MLLMs

A Simple Data Augmentation Strategy for Text-in-Image Scientific VQA

MedCutMix: A Data-Centric Approach to Improve Radiology Vision-Language Pre-training with Disease Awareness

Captioning for Text-Video Retrieval via Dual-Group Direct Preference Optimization

Calibration-Aware Prompt Learning for Medical Vision-Language Models

UnifiedVisual: A Framework for Constructing Unified Vision-Language Datasets

Multilingual Diversity Improves Vision-Language Representations

Decoupling Clinical and Class-Agnostic Features for Reliable Few-Shot Adaptation under Shift

Vision-Language Semantic Aggregation Leveraging Foundation Model for Generalizable Medical Image Segmentation

Aesthetic Image Captioning with Saliency Enhanced MLLMs

Background Matters Too: A Language-Enhanced Adversarial Framework for Person Re-Identification

Linking heterogeneous microstructure informatics with expert characterization knowledge through customized and hybrid vision-language representations for industrial qualification

OwlCap: Harmonizing Motion-Detail for Video Captioning via HMD-270K and Caption Set Equivalence Reward

A CLIP-based Uncertainty Modal Modeling (UMM) Framework for Pedestrian Re-Identification in Autonomous Driving