Unverified50% confidenceFactExact time
JiT (Just image Transformers) is a diffusion generative model that directly predicts clean images.
1
Sources
50%
Confidence
Long-term
Relevance
10/27/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedModern video diffusion models typically adopt U-Net or DiT (Diffusion Transformer) architectures to predict the noise added at each step.66% similarUnverifiedTransformer架构被引入视觉生成（如Vision Transformer、DiT架构），使模型能更好理解图像不同区域间的全局关系，适用于多模态场景中文本token和图像token的统一处理65% similarUnverifiedMeta's Muse Image adopts a Transformer-based masked generation architecture and has a significant advantage in image generation speed.64% similarVerifiedThe Wan (万象) video generation model is based on the DiT (Diffusion Transformer) architecture.63% similarUnverified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63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1102172API
curl https://kongchang.com/api/v1/knowledge/claims/1102172MCP
get_claim(id=1102172)