English
Related papers

Related papers: Fast Autoregressive Video Diffusion and World Mode…

200 papers

Current video captioning methods usually use an encoder-decoder structure to generate text autoregressively. However, autoregressive methods have inherent limitations such as slow generation speed and large cumulative error. Furthermore,…

Computer Vision and Pattern Recognition · Computer Science 2026-04-10 Junbo Wang , Liangyu Fu , Yuke Li , Yining Zhu , Ya Jing , Xuecheng Wu , Jiangbin Zheng

Chunk-wise autoregressive video diffusion models rely on a KV cache of previously generated chunks to avoid redundant computation, but this cache quickly becomes a memory bottleneck as videos grow longer. Methods that quantize the KV cache…

Machine Learning · Computer Science 2026-05-27 Tuna Tuncer , Felix Becker , Thomas Pfeil

Diffusion models have gradually gained prominence in the field of image synthesis, showcasing remarkable generative capabilities. Nevertheless, the slow inference and complex networks, resulting from redundancy at both temporal and…

Computer Vision and Pattern Recognition · Computer Science 2025-03-04 Xuewen Liu , Zhikai Li , Qingyi Gu

Diffusion language models offer parallel token generation and inherent bidirectionality, promising more efficient and powerful sequence modeling compared to autoregressive approaches. However, state-of-the-art diffusion models (e.g., Dream…

Computation and Language · Computer Science 2025-10-10 Zhanqiu Hu , Jian Meng , Yash Akhauri , Mohamed S. Abdelfattah , Jae-sun Seo , Zhiru Zhang , Udit Gupta

Recent large language models (LLMs) are rapidly extending their context windows, yet inference throughput lags due to increasing GPU memory and bandwidth demands. This is because the key-value (KV) cache, an intermediate structure storing…

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…

Computer Vision and Pattern Recognition · Computer Science 2026-03-16 Youngrae Kim , Qixin Hu , C. -C. Jay Kuo , Peter A. Beerel

Video diffusion models have rapidly become the dominant paradigm for high-fidelity generative video synthesis, but their practical deployment remains constrained by severe inference costs. Compared with image generation, video synthesis…

Computer Vision and Pattern Recognition · Computer Science 2026-04-20 Shitong Shao , Lichen Bai , Pengfei Wan , James Kwok , Zeke Xie

Diffusion Transformers (DiT) excel at image and video generation but face computational challenges due to the quadratic complexity of self-attention operators. We propose DiTFastAttn, a post-training compression method to alleviate the…

Computer Vision and Pattern Recognition · Computer Science 2024-10-21 Zhihang Yuan , Hanling Zhang , Pu Lu , Xuefei Ning , Linfeng Zhang , Tianchen Zhao , Shengen Yan , Guohao Dai , Yu Wang

Scaling video diffusion transformers (DiTs) is limited by their quadratic 3D attention, even though most of the attention mass concentrates on a small subset of positions. We turn this observation into VSA, a trainable, hardware-efficient…

Computer Vision and Pattern Recognition · Computer Science 2025-10-29 Peiyuan Zhang , Yongqi Chen , Haofeng Huang , Will Lin , Zhengzhong Liu , Ion Stoica , Eric Xing , Hao Zhang

In this work, we propose Causal Autoregressive Diffusion (CARD), a novel framework that unifies the training efficiency of ARMs with the high-throughput inference of diffusion models. CARD reformulates the diffusion process within a…

Computation and Language · Computer Science 2026-01-30 Junhao Ruan , Bei Li , Yongjing Yin , Pengcheng Huang , Xin Chen , Jingang Wang , Xunliang Cai , Tong Xiao , JingBo Zhu

Diffusion models have significantly advanced video super-resolution (VSR) by enhancing perceptual quality, largely through elaborately designed temporal modeling to ensure inter-frame consistency. However, existing methods usually suffer…

Computer Vision and Pattern Recognition · Computer Science 2025-06-11 Xijun Wang , Xin Li , Bingchen Li , Zhibo Chen

Autoregressive video diffusion models enable open-ended generation through local attention and KV caching. However, existing training-free long-video optimization methods mainly focus on stable extension under a single prompt, making them…

Computer Vision and Pattern Recognition · Computer Science 2026-05-18 Mingqiang Wu , Weilun Feng , Zhefeng Zhang , Haotong Qin , Yuqi Li , Guoxin Fan , Xiaokun Liu , Zhulin An , Libo Huang , Yongjun Xu , Chuanguang Yang

Visual Autoregressive (VAR) has emerged as a promising approach in image generation, offering competitive potential and performance comparable to diffusion-based models. However, current AR-based visual generation models require substantial…

Computer Vision and Pattern Recognition · Computer Science 2024-11-27 Rui Xie , Tianchen Zhao , Zhihang Yuan , Rui Wan , Wenxi Gao , Zhenhua Zhu , Xuefei Ning , Yu Wang

Diffusion Transformers (DiTs) dominate video generation but their high computational cost severely limits real-world applicability, usually requiring tens of minutes to generate a few seconds of video even on high-performance GPUs. This…

Computer Vision and Pattern Recognition · Computer Science 2025-04-29 Haocheng Xi , Shuo Yang , Yilong Zhao , Chenfeng Xu , Muyang Li , Xiuyu Li , Yujun Lin , Han Cai , Jintao Zhang , Dacheng Li , Jianfei Chen , Ion Stoica , Kurt Keutzer , Song Han

Current video generation models perform well at single-shot synthesis but struggle with multi-shot videos, facing critical challenges in maintaining character and background consistency across shots and flexibly generating videos of…

Computer Vision and Pattern Recognition · Computer Science 2025-12-15 Xiangyang Luo , Qingyu Li , Xiaokun Liu , Wenyu Qin , Miao Yang , Meng Wang , Pengfei Wan , Di Zhang , Kun Gai , Shao-Lun Huang

Text-to-Video applications receive increasing attention from the public. Among these, diffusion models have emerged as the most prominent approach, offering impressive quality in visual content generation. However, it still suffers from…

Multimedia · Computer Science 2025-01-09 Desen Sun , Henry Tian , Tim Lu , Sihang Liu

Recent advances in autoregressive video diffusion have enabled real-time frame streaming, yet existing solutions still suffer from temporal repetition, drift, and motion deceleration. We find that naively applying StreamingLLM-style…

Computer Vision and Pattern Recognition · Computer Science 2025-12-05 Jung Yi , Wooseok Jang , Paul Hyunbin Cho , Jisu Nam , Heeji Yoon , Seungryong Kim

Diffusion Transformers (DiTs) power high-fidelity video world models but remain computationally expensive due to sequential denoising and costly spatio-temporal attention. Training-free feature caching accelerates inference by reusing…

Computer Vision and Pattern Recognition · Computer Science 2026-03-24 Umair Nawaz , Ahmed Heakl , Ufaq Khan , Abdelrahman Shaker , Salman Khan , Fahad Shahbaz Khan

We present significant extensions to diffusion-based sequence generation models, blurring the line with autoregressive language models. We introduce hyperschedules, which assign distinct noise schedules to individual token positions,…

Machine Learning · Computer Science 2025-10-08 Nima Fathi , Torsten Scholak , Pierre-André Noël

Modern diffusion models, particularly those utilizing a Transformer-based UNet for denoising, rely heavily on self-attention operations to manage complex spatial relationships, thus achieving impressive generation performance. However, this…

Computer Vision and Pattern Recognition · Computer Science 2024-10-18 Songhua Liu , Weihao Yu , Zhenxiong Tan , Xinchao Wang