Seeing What Matters: A Training-Free Self-Guided Framework for Multimodal Detail Perception and Reasoning
Open paper
Benchmark
Model
Rank
Results
visual-question-answering-on-v-bench
SLoFo high-res
–
Accuracy: 61.78
visual-question-answering-vqa-on-seed-seedbench
SLoFo high-res
–
Accuracy: 67.91
Rank counts only results with a code link.