Train models from scratch with non-English data

• **Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding**
• Kolors: Effective Training of Diffusion Model for Photorealistic Text-to-Image Synthesis

Replace multilingual Encoder

• TAIYI-DIFFUSION-XL: ADVANCING BILINGUAL TEXT-TO-IMAGE GENERATION WITH LARGE VISION-LANGUAGE MODEL SUPPORT
• PanGu-Draw: Advancing Resource-Efficient Text-to-Image Synthesis with Time-Decoupled Training and Reusable Coop-Diffusion
• AltDiffusion: A Multilingual Text-to-Image Diffusion Model

Adapter

• MuLan: Adapting Multilingual Diffusion Models for Hundreds of Languages with Negligible Cost
• PEA-Diffusion: Parameter-Efficient Adapter with Knowledge Distillation in non-English Text-to-Image Generation

Evaluation

• Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset