视觉大模型 / VLM - arXivDaily 专题

2510.21794 2025-10-28 cs.CV cs.AI 86%

Token-Level Inference-Time Alignment for Vision-Language Models

Kejia Chen, Jiawen Zhang, Jiacong Hu, Kewei Gao, Jian Lou, Zunlei Feng, Mingli Song

机构 * Zhejiang University（浙江大学）； Sun Yat-sen University（中山大学）

专题命中 VLM训练与架构：vision-language model(title,abstract)；VLM(abstract)；LLaVA(abstract)；分类 cs.CV、cs.AI

2510.22702 2025-10-28 cs.AI cs.CV cs.ET eess.IV 84%

Atlas Urban Index: A VLM-Based Approach for Spatially and Temporally Calibrated Urban Development Monitoring

Mithul Chander, Sai Pragnya Ranga, Prathamesh Mayekar

机构 * Propheus

专题命中 VLM训练与架构：VLM(title,abstract)；vision-language model(abstract)；分类 cs.CV、cs.AI

Comments An abridged version of this paper will be presented at and appear in the Proceedings of ACM IKDD CODS 2025

2510.22571 2025-10-28 cs.CV cs.AI cs.MM 84%

STATUS Bench: A Rigorous Benchmark for Evaluating Object State Understanding in Vision-Language Models

Mahiro Ukai, Shuhei Kurita, Nakamasa Inoue

机构 * Institute of Science Tokyo（东京科学研究所）； National Institute of Informatics（日本信息处理学会）

专题命中 VLM训练与架构：vision-language model(title,abstract)；VLM(abstract)；分类 cs.CV、cs.AI

2510.21740 2025-10-28 cs.CV cs.AI cs.CL 84%

Diagnosing Bottlenecks in Data Visualization Understanding by Vision-Language Models

Alexa R. Tartaglini, Satchel Grant, Daniel Wurgaft, Christopher Potts, Judith E. Fan

机构 * Department of Computer Science（计算机科学系）； Department of Psychology（心理学系）； Department of Linguistics（语言学系）； Stanford University（斯坦福大学）

专题命中 VLM训练与架构：vision-language model(title,abstract)；VLM(abstract)；分类 cs.CV、cs.AI

2510.21757 2025-10-28 cs.CV 83%

Agro-Consensus: Semantic Self-Consistency in Vision-Language Models for Crop Disease Management in Developing Countries

Mihir Gupta, Pratik Desai, Ross Greer

机构 * The Harker School（哈克尔学校）； University of California, Merced（加州大学默塞德分校）

专题命中 VLM训练与架构：vision-language model(title,abstract)；VLM(abstract)；分类 cs.CV

2505.23566 2025-10-28 cs.CV 83%

Uni-MuMER: Unified Multi-Task Fine-Tuning of Vision-Language Model for Handwritten Mathematical Expression Recognition

Yu Li, Jin Jiang, Jianhua Zhu, Shuai Peng, Baole Wei, Yuxuan Zhou, Liangcai Gao

机构 * Wangxuan Institute of Computer Technology, Peking University（计算机技术研究院，北京大学）

专题命中 VLM训练与架构：vision-language model(title,abstract)；VLM(abstract)；分类 cs.CV

Comments Accepted by NeurIPS 2025 as a spotlight

2510.22641 2025-10-28 cs.LG cs.AI 81%

FastVLM: Self-Speculative Decoding for Fast Vision-Language Model Inference

Divya Jyoti Bajpai, Manjesh Kumar Hanawal

机构 * Dept. of IEOR, IIT Bombay（工业工程与运营研究系，印度理工学院班加罗尔）

专题命中 VLM训练与架构：vision-language model(title,abstract)；分类 cs.AI、cs.LG

Comments Accepted for presentation at the main Conference IJCNLP-AACL 2025

2510.22127 2025-10-28 cs.CV cs.LG 81%

Mint: A Simple Test-Time Adaptation of Vision-Language Models against Common Corruptions

Wenxuan Bao, Ruxi Deng, Jingrui He

机构 * University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）

专题命中 VLM训练与架构：vision-language model(title,abstract)；分类 cs.CV、cs.LG

Comments Accepted by NeurIPS 2025

2510.21879 2025-10-28 cs.CV cs.AI 76%

TernaryCLIP: Efficiently Compressing Vision-Language Models with Ternary Weights and Distilled Knowledge

Shu-Hao Zhang, Wei-Cheng Tang, Chen Wu, Peng Hu, Nan Li, Liang-Jie Zhang, Qi Zhang, Shao-Qun Zhang

机构 * State Key Laboratory of Novel Software Technology, Nanjing University（新型软件技术国家重点实验室）； Microsoft AI（微软人工智能）

专题命中 VLM训练与架构：vision-language model(title)；分类 cs.CV、cs.AI

2510.22370 2025-10-28 cs.RO cs.AI cs.CV cs.LG cs.SE 75%

BLIP-FusePPO: A Vision-Language Deep Reinforcement Learning Framework for Lane Keeping in Autonomous Vehicles

Seyed Ahmad Hosseini Miangoleh, Amin Jalal Aghdasian, Farzaneh Abdollahi

机构 * Department of Electrical Engineering, Amirkabir University of Technology (Tehran Polytechnic)（电气工程系，阿米尔卡比尔技术大学（德黑兰理工大学））

专题命中 VLM训练与架构：vision-language model(abstract)；VLM(abstract)；分类 cs.CV、cs.AI、cs.LG

Comments https://github.com/Amin-A96/BLIP-FusePPO-A-Vision-Language-Deep-Reinforcement-Learning-Framework-for-Lane-Keeping-in-Autonomous.git

2510.22443 2025-10-28 cs.CV cs.LG 73%

Benchmarking Egocentric Multimodal Goal Inference for Assistive Wearable Agents

Vijay Veerabadran, Fanyi Xiao, Nitin Kamra, Pedro Matias, Joy Chen, Caley Drooff, Brett D Roads, Riley Williams, Ethan Henderson, Xuanyi Zhao, Kevin Carlberg, Joseph Tighe, Karl Ridgeway

机构 * Meta Reality Labs（Meta 现实实验室）； Meta FAIR

专题命中 VLM训练与架构：vision-language model(abstract)；VLM(abstract)；分类 cs.CV、cs.LG

Comments Accepted as a spotlight paper at the 39th Conference on Neural Information Processing Systems (NeurIPS 2025)

2510.23203 2025-10-28 cs.CV 70%

DecoDINO: 3D Human-Scene Contact Prediction with Semantic Classification

Lukas Bierling, Davide Pasero, Fleur Dolmans, Helia Ghasemi, Angelo Broere

机构 * University of Amsterdam（阿姆斯特丹大学）

专题命中 VLM训练与架构：vision-language model(abstract)；VLM(abstract)；分类 cs.CV

2510.20819 2025-10-28 cs.CV cs.AI cs.LG 67%

Towards General Modality Translation with Contrastive and Predictive Latent Diffusion Bridge

Nimrod Berman, Omkar Joglekar, Eitan Kosman, Dotan Di Castro, Omri Azencot

机构 * Bosch AI Center（博世人工智能中心）； Ben-Gurion University of the Negev（巴伊兰大学）； Technical University of Munich（慕尼黑技术大学）

专题命中 VLM训练与架构：grounding(abstract)；分类 cs.CV、cs.AI、cs.LG

Comments Accepted as a poster at NeurIPS 2025

2510.21835 2025-10-28 cs.LG cs.AI cs.CL cs.CV 67%

A Multimodal, Multitask System for Generating E Commerce Text Listings from Images

Nayan Kumar Singh

专题命中 VLM训练与架构：VLM(abstract)；分类 cs.CV、cs.AI、cs.LG

Comments 24 pages, 10 figures, 11 tables. Code can be found at: https://github.com/SinghNayanKumar/multimodal-product-lister/

2510.21808 2025-10-28 cs.CV cs.AI 62%

Semantic Relation-Enhanced CLIP Adapter for Domain Adaptive Zero-Shot Learning

Jiaao Yu, Mingjie Han, Jinkun Jiang, Junyu Dong, Tao Gong, Man Lan

机构 * School of Computer Science and Technology, East China Normal University, China（东华大学计算机科学与技术学院）； College of Computer Science and Technology, Ocean University of China, China（中国海洋大学计算机科学与技术学院）

专题命中 VLM训练与架构：vision-language model(abstract)；分类 cs.CV、cs.AI

Comments 5 pages

2510.21806 2025-10-28 cs.CV cs.AI 62%

Frame-Difference Guided Dynamic Region Perception for CLIP Adaptation in Text-Video Retrieval

Jiaao Yu, Mingjie Han, Tao Gong, Jian Zhang, Man Lan

机构 * School of Computer Science and Technology, East China Normal University, China（上海师范大学计算机科学与技术学院）； School of Information Science and Technology, University of Science and Technology of China（中国科学技术大学信息科学与技术学院）

专题命中 VLM训练与架构：vision-language model(abstract)；分类 cs.CV、cs.AI

Comments 5 pages

2506.08010 2025-10-28 cs.CV cs.AI 62%

Vision Transformers Don't Need Trained Registers

Nick Jiang, Amil Dravid, Alexei Efros, Yossi Gandelsman

机构 * UC Berkeley（伯克利大学）

专题命中 VLM训练与架构：vision-language model(abstract)；分类 cs.CV、cs.AI

Comments Project page and code: https://avdravid.github.io/test-time-registers. Accepted to NeurIPS '25 (spotlight)

2411.16503 2025-10-28 cs.CV 57%

Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis

Boming Miao, Chunxiao Li, Xiaoxiao Wang, Andi Zhang, Rui Sun, Zizhe Wang, Yao Zhu

机构 * Beijing Normal University（北京师范大学）； University of Chinese Academy of Sciences（中国科学院大学）； University of Manchester（曼彻斯特大学）； The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））； Tsinghua University（清华大学）

专题命中 VLM训练与架构：vision-language model(abstract)；分类 cs.CV

Comments Updated author formatting; no substantive changes

2510.22994 2025-10-28 cs.CV 57%

SceneDecorator: Towards Scene-Oriented Story Generation with Scene Planning and Scene Consistency

Quanjian Song, Donghao Zhou, Jingyu Lin, Fei Shen, Jiaze Wang, Xiaowei Hu, Cunjian Chen, Pheng-Ann Heng

机构 * Monash University（莫纳什大学）； The Chinese University of Hong Kong（香港中文大学）； National University of Singapore（新加坡国立大学）； South China University of Technology（华南理工大学）

专题命中 VLM训练与架构：VLM(abstract)；分类 cs.CV

Comments Accepted by NeurIPS 2025; Project Page: https://lulupig12138.github.io/SceneDecorator

2506.21864 2025-10-28 cs.CL cs.AI 57%

DeepOmni: Towards Seamless and Smart Speech Interaction with Adaptive Modality-Specific MoE

Hang Shao, Heting Gao, Yunhang Shen, Jiawei Chen, Zuwei Long, Dong Yang, Ke Li, Xing Sun

机构 * Tencent（腾讯）； Fudan University（复旦大学）

专题命中 VLM训练与架构：multimodal large language model(abstract)；分类 cs.AI

Comments Under Review

2503.16188 2025-10-28 cs.CV 57%

Think or Not Think: A Study of Explicit Thinking in Rule-Based Visual Reinforcement Fine-Tuning

Ming Li, Jike Zhong, Shitian Zhao, Yuxiang Lai, Haoquan Zhang, Wang Bill Zhu, Kaipeng Zhang

机构 * Shanghai AI Laboratory（上海人工智能实验室）； University of Southern California（南加州大学）； Emory University（埃默里大学）； Chinese University of Hong Kong（香港中文大学）

专题命中 VLM训练与架构：MLLM(abstract)；分类 cs.CV

Comments Neurips 2025 Spotlight

2502.20625 2025-10-28 cs.CV 57%

T2ICount: Enhancing Cross-modal Understanding for Zero-Shot Counting

Yifei Qian, Zhongliang Guo, Bowen Deng, Chun Tong Lei, Shuai Zhao, Chun Pong Lau, Xiaopeng Hong, Michael P. Pound

机构 * University of Nottingham（诺丁汉大学）； University of St Andrews（圣安德鲁大学）； City University of Hong Kong（香港城市大学）； Nanyang Technology University（南洋理工大学）； Harbin Institute of Technology（哈尔滨工业大学）

专题命中 VLM训练与架构：vision-language model(abstract)；分类 cs.CV

Comments Accepted by CVPR2025

2510.21817 2025-10-28 cs.RO cs.CL cs.LG 57%

VITA-E: Natural Embodied Interaction with Concurrent Seeing, Hearing, Speaking, and Acting

Xiaoyu Liu, Chaoyou Fu, Chi Yan, Chu Wu, Haihan Gao, Yi-Fan Zhang, Shaoqi Dong, Cheng Qian, Bin Luo, Xiuyong Yang, Guanwu Li, Yusheng Cai, Yunhang Shen, Deqiang Jiang, Haoyu Cao, Xing Sun, Caifeng Shan, Ran He

机构 * Nanjing University（南京大学）； Tencent Youtu Lab（腾讯云图实验室）； CASIA（中国科学院自动化研究所）； Fourier Intelligence Inc.（傅里叶智能公司）

专题命中 VLM训练与架构：VLM(abstract)；分类 cs.LG

Comments Homepage: https://lxysl.github.io/VITA-E/

2505.07019 2025-10-28 cs.CV 57%

A Vision-Language Foundation Model for Leaf Disease Identification

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

专题命中 VLM训练与架构：vision-language model(abstract)；分类 cs.CV

1. VLM训练与架构 24 篇

Token-Level Inference-Time Alignment for Vision-Language Models

Atlas Urban Index: A VLM-Based Approach for Spatially and Temporally Calibrated Urban Development Monitoring

STATUS Bench: A Rigorous Benchmark for Evaluating Object State Understanding in Vision-Language Models

Diagnosing Bottlenecks in Data Visualization Understanding by Vision-Language Models

Agro-Consensus: Semantic Self-Consistency in Vision-Language Models for Crop Disease Management in Developing Countries

Uni-MuMER: Unified Multi-Task Fine-Tuning of Vision-Language Model for Handwritten Mathematical Expression Recognition

FastVLM: Self-Speculative Decoding for Fast Vision-Language Model Inference

Mint: A Simple Test-Time Adaptation of Vision-Language Models against Common Corruptions

TernaryCLIP: Efficiently Compressing Vision-Language Models with Ternary Weights and Distilled Knowledge

BLIP-FusePPO: A Vision-Language Deep Reinforcement Learning Framework for Lane Keeping in Autonomous Vehicles

Benchmarking Egocentric Multimodal Goal Inference for Assistive Wearable Agents

DecoDINO: 3D Human-Scene Contact Prediction with Semantic Classification

Towards General Modality Translation with Contrastive and Predictive Latent Diffusion Bridge

A Multimodal, Multitask System for Generating E Commerce Text Listings from Images

Semantic Relation-Enhanced CLIP Adapter for Domain Adaptive Zero-Shot Learning

Frame-Difference Guided Dynamic Region Perception for CLIP Adaptation in Text-Video Retrieval

Vision Transformers Don't Need Trained Registers

Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis

SceneDecorator: Towards Scene-Oriented Story Generation with Scene Planning and Scene Consistency

DeepOmni: Towards Seamless and Smart Speech Interaction with Adaptive Modality-Specific MoE

Think or Not Think: A Study of Explicit Thinking in Rule-Based Visual Reinforcement Fine-Tuning

T2ICount: Enhancing Cross-modal Understanding for Zero-Shot Counting

VITA-E: Natural Embodied Interaction with Concurrent Seeing, Hearing, Speaking, and Acting

A Vision-Language Foundation Model for Leaf Disease Identification