中文
相关论文

相关论文: Rolling Sink: Bridging Limited-Horizon Training an…

200 篇论文

The generation of temporally consistent, high-fidelity driving videos over extended horizons presents a fundamental challenge in autonomous driving world modeling. Existing approaches often suffer from error accumulation and feature…

计算机视觉与模式识别 · 计算机科学 2025-09-03 Jiamin Wang , Yichen Yao , Xiang Feng , Hang Wu , Yaming Wang , Qingqiu Huang , Yuexin Ma , Xinge Zhu

Long-context video modeling is essential for enabling generative models to function as world simulators, as they must maintain temporal coherence over extended time spans. However, most existing models are trained on short clips, limiting…

计算机视觉与模式识别 · 计算机科学 2025-05-20 Yuchao Gu , Weijia Mao , Mike Zheng Shou

Recent approaches to real-time long video generation typically employ streaming tuning strategies, attempting to train a long-context student using a short-context (memoryless) teacher. In these frameworks, the student performs long…

计算机视觉与模式识别 · 计算机科学 2026-02-06 Shuo Chen , Cong Wei , Sun Sun , Ping Nie , Kai Zhou , Ge Zhang , Ming-Hsuan Yang , Wenhu Chen

Distilled autoregressive diffusion models facilitate real-time short video synthesis but suffer from severe error accumulation during long-sequence generation. While existing Test-Time Optimization (TTO) methods prove effective for images…

计算机视觉与模式识别 · 计算机科学 2026-03-11 Xunzhi Xiang , Zixuan Duan , Guiyu Zhang , Haiyu Zhang , Zhe Gao , Junta Wu , Shaofeng Zhang , Tengfei Wang , Qi Fan , Chunchao Guo

Video diffusion models perform well in short-video synthesis, but their training-free extension to long videos often suffers from content drift, temporal inconsistency, and over-smoothed dynamics. Existing methods improve temporal…

计算机视觉与模式识别 · 计算机科学 2026-05-08 Fangda Chen , Shanshan Zhao , Longrong Yang , Chuanfu Xu , Zhigang Luo , Long Lan

Distilled autoregressive (AR) video models enable efficient streaming generation but frequently misalign with human visual preferences. Existing reinforcement learning (RL) frameworks are not naturally suited to these architectures,…

计算机视觉与模式识别 · 计算机科学 2026-03-19 Songchun Zhang , Zeyue Xue , Siming Fu , Jie Huang , Xianghao Kong , Y Ma , Haoyang Huang , Nan Duan , Anyi Rao

Autoregressive (AR) diffusion enables streaming, interactive long-video generation by producing frames causally, yet maintaining coherence over minute-scale horizons remains challenging due to accumulated errors, motion drift, and content…

计算机视觉与模式识别 · 计算机科学 2025-12-05 Yifei Yu , Xiaoshan Wu , Xinting Hu , Tao Hu , Yangtian Sun , Xiaoyang Lyu , Bo Wang , Lin Ma , Yuewen Ma , Zhongrui Wang , Xiaojuan Qi

Long video generation remains a challenging and compelling topic in computer vision. Diffusion based models, among the various approaches to video generation, have achieved state of the art quality with their iterative denoising procedures.…

计算机视觉与模式识别 · 计算机科学 2025-03-14 Siyang Zhang , Harry Yang , Ser-Nam Lim

We address the problem of generating long-horizon videos for robotic manipulation tasks. Text-to-video diffusion models have made significant progress in photorealism, language understanding, and motion generation but struggle with…

计算机视觉与模式识别 · 计算机科学 2025-06-30 Liudi Yang , Yang Bai , George Eskandar , Fengyi Shen , Mohammad Altillawi , Dong Chen , Soumajit Majumder , Ziyuan Liu , Gitta Kutyniok , Abhinav Valada

We propose LayerSync, a domain-agnostic approach for improving the generation quality and the training efficiency of diffusion models. Prior studies have highlighted the connection between the quality of generation and the representations…

计算机视觉与模式识别 · 计算机科学 2026-02-20 Yasaman Haghighi , Bastien van Delft , Mariam Hassan , Alexandre Alahi

Video prediction is an important yet challenging problem; burdened with the tasks of generating future frames and learning environment dynamics. Recently, autoregressive latent video models have proved to be a powerful video prediction…

计算机视觉与模式识别 · 计算机科学 2022-09-16 Younggyo Seo , Kimin Lee , Fangchen Liu , Stephen James , Pieter Abbeel

Autoregressive video diffusion models support real-time synthesis but suffer from error accumulation and context loss over long horizons. We discover that attention heads in AR video diffusion transformers serve functionally distinct roles…

计算机视觉与模式识别 · 计算机科学 2026-05-15 Jiahao Tian , Yiwei Wang , Gang Yu , Chi Zhang

Long-horizon action-conditioned video generation aims to synthesize temporally coherent videos that follow complex action instructions over extended horizons, requiring procedural ordering, persistent action execution, and scene consistency…

计算机视觉与模式识别 · 计算机科学 2026-05-22 Yu Yang , Yue Liao , Jianbiao Mei , Baisen Wang , Xuemeng Yang , Licheng Wen , Jiangning Zhang , Xiangtai Li , Liang Lv , Hanlin Chen , Botian Shi , Yong Liu , Shuicheng Yan , Gim Hee Lee

Autoregressive video diffusion models have emerged as a scalable paradigm for long video generation. However, they often suffer from severe extrapolation failure, where rapid error accumulation leads to significant temporal degradation when…

计算机视觉与模式识别 · 计算机科学 2026-03-16 Jia Li , Xiaomeng Fu , Xurui Peng , Weifeng Chen , Youwei Zheng , Tianyu Zhao , Jiexi Wang , Fangmin Chen , Xing Wang , Hayden Kwok-Hay So

Autoregressive diffusion enables real-time frame streaming, yet existing sliding-window caches discard past context, causing fidelity degradation, identity drift, and motion stagnation over long horizons. Current approaches preserve a fixed…

计算机视觉与模式识别 · 计算机科学 2026-03-16 Youngrae Kim , Qixin Hu , C. -C. Jay Kuo , Peter A. Beerel

Real-time motion-controllable video generation remains challenging due to the inherent latency of bidirectional diffusion models and the lack of effective autoregressive (AR) approaches. Existing AR video diffusion models are limited to…

计算机视觉与模式识别 · 计算机科学 2026-03-10 Kesen Zhao , Jiaxin Shi , Beier Zhu , Junbao Zhou , Xiaolong Shen , Yuan Zhou , Qianru Sun , Hanwang Zhang

Current frontier video diffusion models have demonstrated remarkable results at generating high-quality videos. However, they can only generate short video clips, normally around 10 seconds or 240 frames, due to computation limitations…

计算机视觉与模式识别 · 计算机科学 2025-05-20 Desai Xie , Zhan Xu , Yicong Hong , Hao Tan , Difan Liu , Feng Liu , Arie Kaufman , Yang Zhou

The quadratic time and memory complexity of the attention mechanism in modern Transformer based video generators makes end-to-end training for ultra high resolution videos prohibitively expensive. Motivated by this limitation, we introduce…

计算机视觉与模式识别 · 计算机科学 2025-12-02 Yunfeng Wu , Jiayi Song , Zhenxiong Tan , Zihao He , Songhua Liu

Recent advancements in image relighting models, driven by large-scale datasets and pre-trained diffusion models, have enabled the imposition of consistent lighting. However, video relighting still lags, primarily due to the excessive…

计算机视觉与模式识别 · 计算机科学 2025-03-13 Yujie Zhou , Jiazi Bu , Pengyang Ling , Pan Zhang , Tong Wu , Qidong Huang , Jinsong Li , Xiaoyi Dong , Yuhang Zang , Yuhang Cao , Anyi Rao , Jiaqi Wang , Li Niu

Reinforcement learning for training end-to-end autonomous driving models in closed-loop simulations is gaining growing attention. However, most simulation environments differ significantly from real-world conditions, creating a substantial…

计算机视觉与模式识别 · 计算机科学 2025-08-22 Chaojun Ni , Guosheng Zhao , Xiaofeng Wang , Zheng Zhu , Wenkang Qin , Xinze Chen , Guanghong Jia , Guan Huang , Wenjun Mei