视频大模型

Perceive, Reflect and Understand Long Video: Progressive Multi-Granular Clue Exploration with Interactive Agents

Jiahua Li, Kun Wei, Zhe Xu, Zibo Su, Xu Yang, Cheng Deng

机构 * School of Electronic Engineering, Xidian University（西安电子科技大学电子工程学院） ； Department of Computer Science and Engineering, Hong Kong University of Science and Technology（香港理工大学计算机科学与工程系） ； College of Computer and Information, Hohai University（河海大学计算机与信息学院） 纠错

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

Virtual Fitting Room: Generating Arbitrarily Long Videos of Virtual Try-On from a Single Image -- Technical Preview

Jun-Kun Chen, Aayush Bansal, Minh Phuoc Vo, Yu-Xiong Wang

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

专题命中 长视频与时序推理 ：long video(title,abstract)；video generation(abstract)；分类 cs.CV

Comments Project Page: https://immortalco.github.io/VirtualFittingRoom/

StreamMem: Query-Agnostic KV Cache Memory for Streaming Video Understanding

Yanlai Yang, Zhuokai Zhao, Satya Narayan Shukla, Aashu Singh, Shlok Kumar Mishra, Lizhu Zhang, Mengye Ren

机构 * Meta AI ； New York University（纽约大学） 纠错

专题命中 长视频与时序推理 ：video understanding(title,abstract)；long video(abstract)；分类 cs.CV

Comments 15 pages, 3 figures

Vamba: Understanding Hour-Long Videos with Hybrid Mamba-Transformers

Weiming Ren, Wentao Ma, Huan Yang, Cong Wei, Ge Zhang, Wenhu Chen

机构 * University of Waterloo（滑铁卢大学） ； University of Toronto（多伦多大学） ； Kuaishou Technology（快手科技） ； Vector Institute（向量研究所） ； M-A-P 纠错

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

Comments ICCV 2025 Camera Ready Version. Project Page: https://tiger-ai-lab.github.io/Vamba/

Video Depth Anything: Consistent Depth Estimation for Super-Long Videos

Sili Chen, Hengkai Guo, Shengnan Zhu, Feihu Zhang, Zilong Huang, Jiashi Feng, Bingyi Kang

机构 * ByteDance（字节跳动） 纠错

专题命中 长视频与时序推理 ：long video(title,abstract)；video generation(abstract)；分类 cs.CV

Comments Project page: https://videodepthanything.github.io/

Threading Keyframe with Narratives: MLLMs as Strong Long Video Comprehenders

Bo Fang, Wenhao Wu, Qiangqiang Wu, Yuxin Song, Antoni B. Chan

机构 * Department of Computer Science, City University of Hong Kong（香港城市大学计算机科学系） ； Baidu Inc.（百度公司） ； University of Sydney（悉尼大学） 纠错

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

ViSMaP: Unsupervised Hour-long Video Summarisation by Meta-Prompting

Jian Hu, Dimitrios Korkinof, Shaogang Gong, Mariano Beguerisse-Diaz

机构 * Queen Mary University of London（伦敦玛丽女王大学） ； Spotify 纠错

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

AdaVid: Adaptive Video-Language Pretraining

Chaitanya Patel, Juan Carlos Niebles, Ehsan Adeli

专题命中 长视频与时序推理 ：video-language(title,abstract)；long video(abstract)；分类 cs.CV

Comments CVPRW 2025. Project Page: https://chaitanya100100.github.io/AdaVid/

Loong: Generating Minute-level Long Videos with Autoregressive Language Models

Yuqing Wang, Tianwei Xiong, Daquan Zhou, Zhijie Lin, Yang Zhao, Bingyi Kang, Jiashi Feng, Xihui Liu

专题命中 长视频与时序推理 ：long video(title,abstract)；video generation(abstract)；分类 cs.CV

Comments Project page: https://yuqingwang1029.github.io/Loong-video

Does Your Vision-Language Model Get Lost in the Long Video Sampling Dilemma?

Tianyuan Qu, Longxiang Tang, Bohao Peng, Senqiao Yang, Bei Yu, Jiaya Jia

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

ReWind: Understanding Long Videos with Instructed Learnable Memory

Anxhelo Diko, Tinghuai Wang, Wassim Swaileh, Shiyan Sun, Ioannis Patras

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

MTV-Inpaint: Multi-Task Long Video Inpainting

Shiyuan Yang, Zheng Gu, Liang Hou, Xin Tao, Pengfei Wan, Xiaodong Chen, Jing Liao

专题命中 长视频与时序推理 ：long video(title,abstract)；text-to-video(abstract)；分类 cs.CV

LongVILA: Scaling Long-Context Visual Language Models for Long Videos

Yukang Chen, Fuzhao Xue, Dacheng Li, Qinghao Hu, Ligeng Zhu, Xiuyu Li, Yunhao Fang, Haotian Tang, Shang Yang, Zhijian Liu, Ethan He, Hongxu Yin, Pavlo Molchanov, Jan Kautz, Linxi Fan, Yuke Zhu, Yao Lu, Song Han

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

Comments Code and models are available at https://github.com/NVlabs/VILA/tree/main/longvila

SAVEn-Vid: Synergistic Audio-Visual Integration for Enhanced Understanding in Long Video Context

Jungang Li, Sicheng Tao, Yibo Yan, Xiaojie Gu, Haodong Xu, Xu Zheng, Yuanhuiyi Lyu, Linfeng Zhang, Xuming Hu

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

Comments The publication has some processing errors (language short-cuts in synthetic data are not avoided) that invalidate some of the conclusions

A Video is Worth 10,000 Words: Training and Benchmarking with Diverse Captions for Better Long Video Retrieval

Matthew Gwilliam, Michael Cogswell, Meng Ye, Karan Sikka, Abhinav Shrivastava, Ajay Divakaran

专题命中 长视频与时序推理 ：long video(title,abstract)；video language model(abstract)；分类 cs.CV

Comments 17 pages, 16 tables, 8 figures. To appear at WACV 2025

Towards Chunk-Wise Generation for Long Videos

Siyang Zhang, Ser-Nam Lim

专题命中 长视频与时序推理 ：long video(title,abstract)；video generation(abstract)；分类 cs.CV

Goldfish: Vision-Language Understanding of Arbitrarily Long Videos

Kirolos Ataallah, Xiaoqian Shen, Eslam Abdelrahman, Essam Sleiman, Mingchen Zhuge, Jian Ding, Deyao Zhu, Jürgen Schmidhuber, Mohamed Elhoseiny

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

Comments 25 pages, 11 figures, accepted by ECCV 2024

Koala: Key frame-conditioned long video-LLM

Reuben Tan, Ximeng Sun, Ping Hu, Jui-hsien Wang, Hanieh Deilamsalehy, Bryan A. Plummer, Bryan Russell, Kate Saenko

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

Comments Accepted at CVPR 2024 as a poster highlight

MovieChat+: Question-aware Sparse Memory for Long Video Question Answering

Enxin Song, Wenhao Chai, Tian Ye, Jenq-Neng Hwang, Xi Li, Gaoang Wang

专题命中 长视频与时序推理 ：long video(title,abstract)；video understanding(abstract)；分类 cs.CV

LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment

Bin Zhu, Bin Lin, Munan Ning, Yang Yan, Jiaxi Cui, HongFa Wang, Yatian Pang, Wenhao Jiang, Junwu Zhang, Zongwei Li, Wancai Zhang, Zhifeng Li, Wei Liu, Li Yuan

专题命中 长视频与时序推理 ：video-language(title,abstract)；long video(abstract)；分类 cs.CV

Comments Accepted by ICLR 2024

Towards End-to-End Generative Modeling of Long Videos with Memory-Efficient Bidirectional Transformers

Jaehoon Yoo, Semin Kim, Doyup Lee, Chiheon Kim, Seunghoon Hong

专题命中 长视频与时序推理 ：long video(title,abstract)；video generation(abstract)；分类 cs.CV

Test of Time: Instilling Video-Language Models with a Sense of Time

Piyush Bagad, Makarand Tapaswi, Cees G. M. Snoek

专题命中 长视频与时序推理 ：video-language(title,abstract)；video understanding(abstract)；分类 cs.CV

Comments Accepted for publication at CVPR 2023. Project page: https://bpiyush.github.io/testoftime-website/index.html

PyraVid: Hierarchical Multimodal Memory for Long-Horizon Video Reasoning

PyraVid: 用于长时间视频推理的分层多模态记忆

Sikuan Yan, Sicheng Dong, Haotong Wang, Ercong Nie, Yilun Liu, Jinhe Bi, Yingjie Xu, Susanna Schwarzmann, Riccardo Trivisonno, Volker Tresp, Yunpu Ma

专题命中 长视频与时序推理 ：video reasoning(title)；video understanding(abstract)；long video(abstract)

AI总结 本文提出PyraVid，一种分层多模态记忆框架，通过事件分割理论启发，解决长视频中多模态信息整合、人本信息对齐和证据聚合等挑战，提升长时间视频推理性能。

NARU: A Benchmark for NARrative Evolution and Cultural Nuance Understanding in Japanese Extreme Long Video

NARU：用于理解日语超长视频中叙事演变与文化细微差别的基准

Yuheng Huang, Jianlang Chen, Jiayang Song, Hua Qi, Aza Kai, Vincent Markert, Edison Marrese-Taylor, Jianjun Zhao, Lei Ma

机构 * The University of Tokyo（东京大学） ； Kyushu University（九州大学） ； Macau University of Science and Technology（澳门科技大学） ； Infinimind Japan Inc.（Infinimind日本公司） ； University of Alberta（阿尔伯塔大学） 纠错

专题命中 长视频与时序推理 ：long video(title)；video understanding(abstract)；分类 cs.CV、cs.MM

AI总结 该研究推出NARU基准，涵盖155个146.8小时日语视频的1481个问题，评估模型在长程叙事整合与文化推理上的局限，为MLLM开发提供测试平台。

Comments Yuheng Huang and Jianlang Chen contributed equally to this work. More details available on the project's website https://ma-labo.github.io/naru/ and https://infinimind.io/en/company/news/2026/narubench-release

TTOM: Test-Time Optimization and Memorization for Compositional Video Generation

TTOM：测试时优化与记忆化用于组合视频生成

Leigang Qu, Ziyang Wang, Na Zheng, Wenjie Wang, Liqiang Nie, Tat-Seng Chua

机构 * National University of Singapore（国立新加坡大学） ； University of Science and Technology of China（中国科学技术大学） ； Harbin Institute of Technology (Shenzhen)（哈尔滨工业大学（深圳）） 纠错

专题命中 长视频与时序推理 ：video generation(title,abstract)；分类 cs.CV、cs.MM

AI总结 TTOM通过测试时优化与记忆化机制，提升组合视频生成的跨模态对齐能力，实现高效且可扩展的实时生成。

Comments ICLR 2026 Camera-ready. Project page: https://ttom-t2v.github.io/

HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training

Qinghao Ye, Guohai Xu, Ming Yan, Haiyang Xu, Qi Qian, Ji Zhang, Fei Huang

专题命中 长视频与时序推理 ：video-language(title,abstract)；分类 cs.CV、cs.MM

Shallow Optical Flow Three-Stream CNN for Macro- and Micro-Expression Spotting from Long Videos

Gen-Bing Liong, John See, Lai-Kuan Wong

专题命中 长视频与时序推理 ：long video(title,abstract)；分类 cs.CV、cs.MM

Comments Accepted for publication in ICIP2021. 9 pages, including 3 pages of supplemental notes

Spotting Macro- and Micro-expression Intervals in Long Video Sequences

Ying He, Su-Jing Wang, Jingting Li, Moi Hoon Yap

专题命中 长视频与时序推理 ：long video(title,abstract)；分类 cs.CV、eess.IV

Comments 7 pages, 4 figures and 3 tables

The Runner-up Solution for YouTube-VIS Long Video Challenge 2022

Junfeng Wu, Yi Jiang, Qihao Liu, Xiang Bai, Song Bai

专题命中 长视频与时序推理 ：long video(title,abstract)；分类 cs.CV

Comments The Runner-up Solution for YouTube-VIS Long Video Challenge 2022, ECCV 2022 Workshop. arXiv admin note: text overlap with arXiv:2207.10661

Lost in Adaptation: Layer-Selective Recovery of Temporal Reasoning in Video-Language Models

层次间推理：通过层选择性融合恢复视频-语言模型中的时间推理

Zihang Fu, Haonan Wang, Jian Kang, Kenji Kawaguchi, Jiaying Wu

机构 * National University of Singapore（新加坡国立大学） ； MBZUAI（穆罕默德·本·扎耶德人工智能大学） 纠错

专题命中 长视频与时序推理 ：video-language(title,abstract)；分类 cs.CV

AI总结 本文提出MERIT框架，通过层选择性融合提升视频-语言模型的时间推理能力，同时保持时间感知能力，优于全模型融合和随机层选择。