面向高效自回归视频生成的运动感知缓存
TLDR
MotionCache基于像素级运动自适应缓存,加速自回归视频生成,实现6.28倍和1.64倍提速且保持质量。
评分理由
The paper presents a novel motion-aware caching method with theoretical grounding and strong empirical speedups on multiple models, which is a clear strength. However, the scope is limited to video generation efficiency and does not address world modeling or related broader claims, weakening relevance to the specified keywords.
Read-first 评分解释
综合优先阅读分 18,由主题、引用、图谱、方法、可复现性和近期性等信号加权得到。 原始总分保留为 2。
研究版图角色
前沿论文
排序敏感性
稳定性:volatile;排名波动范围:59。