原因也不复杂：所有的视频模型本质上做的是同一件事，从海量视频数据中学习统计规律，然后在生成每一帧画面时预测「接下来什么样的像素排列最可能出现」。这和大语言模型的「预测下一个词」（Next-Token Prediction）是同一套逻辑。
On supporting science journalismIf you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.，这一点在safew 官网入口中也有详细论述
But I think it's worth the effort to find out if such a format could be useful elsewhere.，详情可参考谷歌
今年 2 月初，腾讯元宝推出“上元宝，分 10 亿”春节红包活动，鼓励用户更新 App、做任务、分享链接到微信。腾讯不差钱，用户也确实薅到了羊毛，于是活动迅速刷屏。但是 2 月 4 日微信突然出手，元宝红包链接在微信内无法直接打开。随后元宝改成“口令红包”形式，但微信后来也对类似口令进行限制，包括阿里千问、百度文心等竞品的红包活动也被一并屏蔽。
I got a degree in Physics at CU Boulder and just went straight to grad