Architecture option
DiT with diffusion head (DiT^DH)
A DiT variant whose final block is replaced by a small per-token diffusion head, used to decode high-dimensional semantic latents where a single linear projection is insufficient.
Introduced by
- RAE 2025-10
Used by (1)
Alternatives on this axis
Diffusion Transformer (DiT), MM-DiT (dual-stream joint attention), Single-stream (unified-weight) DiT, UNet backbone