GenEval

The 'GenEval' dataset/benchmark is used to evaluate the performance of unified multimodal models in generating images based on reasoning and visual understanding.

Papers using GenEval (3)

Reinforcing the Generation Order of Multimodal Masked Diffusion Models2026OctoT2I: A Self-Evolving Agentic Text-to-Image Router2026Can AI Draw Science? A Benchmark for Evaluating Scientific Figure Generation by Text-to-Image and Multimodal Models2026