ProLaViT: Learning Progressive Latent Visual Thoughts in Structured Latent Space
Open paper
Benchmark
Model
Rank
Results
visual-question-answering-on-v-bench
ProLaViT (Full)
#15
Accuracy: 80.11
visual-question-answering-vqa-on-blink
ProLaViT (Full)
#31
Accuracy: 57.23
Rank counts only results with a code link.