Paying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models
Open paper
Benchmark
Model
Rank
Results
chart-question-answering-on-chartqa
VISE (Qwen3-VL-8B)
#43
1:1 Accuracy: 85.41
multimodal-reasoning-on-mmmu
VISE (Qwen3-VL-8B)
#87
Accuracy: 52.69
Rank counts only results with a code link.