349 citations · 900 across the 26 of their papers we have counts for
1 paper · 2 filters
Yuqing Wang, Tianwei Xiong, Daquan Zhou +5
It is desirable but challenging to generate content-rich long videos in the scale of minutes. Autoregressive large language models (LLMs) have achieved great success in generating…