Prism:Dynamic Sparse Attention for Native 2K Joint Video-Audio Generation Model Training Shuyuan Tu 1, Qi Tian 2, Yinming Huang 1, Yue Wu 2, Xintong Han 2, Kaihang Pan 3, Weijie Kong 2, Jiangfeng Xiong 2, Jian-Wei Zhang 2, Zuxuan Wu 1, Yu-Gang Jiang 1 [ 1 Fudan University; 2 Tencent Hunyuan; 3 Zhejiang University] Native training joint video-audio generation models at higher resolutions empowers them to learn richer visual details and sharper motion dynamics. However, full attention incurs quadratic cost and, as resolution increases, spreads attention over increasingly redundant tokens, diluting learning signals for informative content and disrupting pretrained priors. Existing sparse…