TempoMaster：通过下一帧率预测实现高效长视频生成
TempoMaster: Efficient Long Video Generation via Next-Frame-Rate Prediction
把长视频生成拆成「先低帧率粗蓝图、再逐级升帧率精修」的层级自回归范式，帧内双向注意力保证并行效率、跨帧率自回归保证长程时序一致，可直接迁移到任何需要长时序一致性的视频扩散/流匹配采样流程中。
原文依据 formulates long video generation as next-frame-rate prediction
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出 TempoMaster，一种将长视频生成建模为下一帧率预测的新框架。具体而言，首先生成一个低帧率片段，作为整个视频序列的粗略蓝图，随后逐步提升帧率以精修视觉细节与运动连续性。在生成过程中，TempoMaster 在每个帧率层级内部采用双向注意力，同时在帧率层级之间执行自回归，从而在实现高效并行合成的同时获得长程时序一致性。大量实验表明，TempoMaster 在长视频生成上取得了新的最优性能，在视觉质量与时序质量两方面均表现优异。
英文原文摘要 720 字符
We present TempoMaster, a novel framework that formulates long video generation as next-frame-rate prediction. Specifically, we first generate a low-frame-rate clip that serves as a coarse blueprint of the entire video sequence, and then progressively increase the frame rate to refine visual details and motion continuity. During generation, TempoMaster employs bidirectional attention within each frame-rate level while performing autoregression across frame rates, thus achieving long-range temporal coherence while enabling efficient and parallel synthesis. Extensive experiments demonstrate that TempoMaster establishes a new state-of-the-art in long video generation, excelling in both visual and temporal quality.