Mixture of Specialized Vision Experts: Unlocking Complementary Visual Insights for Faithful MLLM Reasoning
Open paper
Benchmark
Model
Rank
Results
visual-question-answering-vqa-on-mmvp
MoSVE
#2
Paired Accuracy: 57
object-hallucination-evaluation-on-pope
MoSVE
#22
F1: 86.5
Rank counts only results with a code link.