DiffusionVL: Translating Any Autoregressive Models into Diffusion Vision Language Models

Benchmark Model Rank Results
mathematical-reasoning-on-mathverseDiffusionVL#29Accuracy: 36.29
chart-question-answering-on-chartqaDiffusionVL#501:1 Accuracy: 84.2
visual-question-answering-vqa-on-realworldqaDiffusionVL#58Accuracy: 68.0
multimodal-reasoning-on-mathvisionDiffusionVL#60Accuracy: 19.08
multimodal-reasoning-on-mathvistaDiffusionVL#83Accuracy: 64.3
multimodal-reasoning-on-mmmuDiffusionVL#99Accuracy: 49.3