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

Yicheng Ji, Zhizhou Zhong, Jun Zhang, Qin Yang, XiTai Jin, Ying Qin, Wenhan Luo, Shuiyang Mao, Wei Liu, Huan Li
May 2026
Cite arXiv Code Project
Type
Paper-Preprint
Publication
arXiv preprint
Yicheng Ji, Zhizhou Zhong, Jun Zhang, Qin Yang, XiTai Jin, Ying Qin, Wenhan Luo, Shuiyang Mao, Wei Liu, Huan Li