[KongchangAI]
Unverified50% confidenceFactExact time

文生视频模型通常基于扩散变换器(Diffusion Transformer)架构,将运动先验与外观风格解耦处理

1
Sources
50%
Confidence
Long-term
Relevance
7/15/2026
First Seen

Sources

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/517850
API
curl https://kongchang.com/api/v1/knowledge/claims/517850
MCP
get_claim(id=517850)