paper-with-me

Papers

Efficient Autoregressive Video Diffusion with Dummy Head

2026-01-28 · Hang Guo, Zhaoyang Jia, Jiahao Li, Bin Li, Yuanhao Cai, Jiangshan Wang, Yawei Li, Yan Lu arxiv

The autoregressive video diffusion model has recently gained considerable research interest due to its causal modeling and iterative denoising. In this work, we identify that the multi-head self-attention in these models under-utilizes historical frames: approximately 25% heads attend almost exclusively to the current frame, and discarding their KV caches incurs only minor performance degradation. Building upon this, we propose Dummy Forcing, a simple yet effective method to control context accessibility across different heads. Specifically, the proposed heterogeneous memory allocation reduces head-wise context redundancy, accompanied by dynamic head programming to adaptively classify head types. Moreover, we develop a context packing technique to achieve more aggressive cache compression. Without additional training, our Dummy Forcing delivers up to 2.0x speedup over the baseline, supporting video generation at 24.3 FPS with less than 0.5% quality drop. Project page is available at https://csguoh.github.io/project/DummyForcing/.

📄 PDF Abstract BibTeX arXiv:2601.20499

Code (0)

등록된 구현이 없습니다.

Tasks

Video Generation

Similar Papers 제목 키워드 기반

HeadCast: Casting Attention Heads for Efficient Autoregressive Video Generation

2026-07-22 · Jinliang Shen, Lianghao Su, Zheming Li, Kang He 외 arxiv

Autoregressive (AR) video diffusion models have become a promising paradigm for long and streaming video synthesis, but the continuously growing Key-Value (KV) cache makes attention the dominant inference cost, especiall…

Video Generation

Forcing-KV: Hybrid KV Cache Compression for Efficient Autoregressive Video Diffusion Models

2026-05-10 · Yicheng Ji, Zhizhou Zhong, Jun Zhang, Qin Yang 외 arxiv

Autoregressive (AR) video diffusion models adopt a streaming generation framework, enabling long-horizon video generation with real-time responsiveness, as exemplified by the Self Forcing training paradigm. However, exis…

Video Generation

DAWN: Dynamic Frame Avatar with Non-autoregressive Diffusion Framework for Talking Head Video Generation

2024-10-17 · Hanbo Cheng, Limin Lin, Chenyu Liu, Pengcheng Xia 외

Talking head generation intends to produce vivid and realistic talking head videos from a single portrait and speech audio clip. Although significant progress has been made in diffusion-based talking head generation, alm…

Talking Head GenerationVideo Generation

Head Forcing: Long Autoregressive Video Generation via Head Heterogeneity

2026-05-14 · Jiahao Tian, Yiwei Wang, Gang Yu, Chi Zhang arxiv

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 function…

Video Generation

DSA: Dynamic Step Allocation for Fast Autoregressive Video Generation

2026-06-03 · Thanh-Tung Le, Yunhan Zhao, Menglei Chai, Zhengyang Shen 외 arxiv

Video diffusion transformers have achieved state-of-the-art visual quality, but their high inference cost remains a major bottleneck for real-time applications. Recent distillation frameworks produce autoregressive video…

Video Generation