Dynamics-Aware Sparse Attention for Efficient Autoregressive Video Diffusion
论文方法，未开源，无法直接部署
Autoregressive Diffusion Transformers have emerged as a powerful paradigm for long video generation. However, they are often constrained by prohibitive computational overhead and substantial memory footprints. This inefficiency arises from dense computation mechanisms that alloca
判断口径 agnes-3.0-flash · 入库 #61 · 回首页