DPG-Bench

DPG-Bench is a benchmark used to evaluate the performance of models in text-to-image synthesis, specifically assessing their ability to generate images based on text prompts.

Papers using DPG-Bench (3)

Mural: Transferring LLM knowledge to image generation via Mixture-of-Transformers2026Refinement via Regeneration: Enlarging Modification Space Boosts Image Refinement in Unified Multimodal Models2026AlphaGRPO: Unlocking Self-Reflective Multimodal Generation in UMMs via Decompositional Verifiable Reward2026