Meta Inverse Prompting for Video Generative Models
逆向提示研究，未放权重，暂时用不上
Recent advances in video generative models have enabled flexible conditional generation and editing pipelines, such as text-to-video, image-to-video, and reference-guided synthesis. However, these workflows are inherently one-directional, mapping conditions to videos without prov
判断口径 agnes-3.0-flash · 入库 #77 · 回首页