AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

MMFace-DiT: A Dual-Stream Diffusion Transformer for High-Fidelity Multimodal Face Generation

Bharath Krishnamurthy, Ajita Rattani

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Modality Embedder to the baseline DiT (ablation)baseline DiT (SD2 VAE) · Text + Mask face generationFID: 44.52 -> 40.49Improved
add Dual-Stream design (ablation)DiT + Modality Embedder · Text + Mask face generationCLIP score: 24.53 -> 29.69Improved
add Dual-Stream design (ablation)DiT + Modality Embedder · Text + Mask face generationmIoU: 46.34 -> 48.91Improved
replace standard attention with shared RoPE attention (ablation)DiT + ME + Dual-Stream · Text + Mask face generationmIoU: 48.91 -> 50.05Improved
add Modality Embedder (no extra training per modality) (ablation)baseline DiT (SD2 VAE) · Text + Mask face generationCLIP score: 24.53 -> 24.31No clear effect
replace SD2 VAE with Flux VAE (ablation)MMFace-DiT · Text + Mask face generationFID: 33.77 -> 27.95Improved
replace SD2 VAE with Flux VAE (ablation)MMFace-DiT · Text + Mask face generationmIoU: 50.05 -> 49.16Hurt
use SD3 VAE instead of Flux VAE (ablation)MMFace-DiT · Text + Mask face generationFID: 27.95 -> 20.73Improved
use SD3 VAE instead of Flux VAE (ablation)MMFace-DiT · Text + Mask face generationLPIPS: 0.340 -> 0.368Hurt
use Qwen-Image VAE instead of SDXL/SD2 (ablation)MMFace-DiT · Text + Mask face generationFID: 23.00 -> 33.09Hurt
train with rectified flow matching instead of DDPM (ablation)MMFace-DiT · Text + Sketch face generationFID: 27.67 -> 9.14Improved
train with rectified flow matching instead of DDPM (ablation)MMFace-DiT · Text + Mask face generationFID: 27.95 -> 16.63Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy