Unverified50% confidenceFactExact time
Sora、Runway Gen系列和Pika大多基于扩散模型或视频Transformer架构
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
Valid until: 10/10/2026
Sources
AI生成视频的进步与瓶颈：光影惊艳，逻辑存疑
redditr/midjourney7/11/2026
Related Claims
Partially VerifiedSora采用DiT（Diffusion Transformer）架构，将视频压缩为时空patch序列后输入Transformer，以Transformer替代传统U-Net作为去噪骨干网络72% similarUnverifiedSora采用了Diffusion Transformer（DiT）架构，将Transformer全局注意力机制引入扩散框架67% similarUnverified图像生成领域正在经历从U-Net到Transformer的架构范式转移，传统扩散模型如Stable Diffusion 1.x和2.x主要依赖U-Net编码器-解码器结构60% similarUnverifiedSwitch Transformer和Mixtral系列广泛采用了MoE架构59% similarUnverifiedMeta的LLaMA系列、Google的Gemma、Mistral AI的Mistral/Mixtral、阿里的Qwen系列都在Transformers中获得支持59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/491569API
curl https://kongchang.com/api/v1/knowledge/claims/491569MCP
get_claim(id=491569)