Forest Before Trees: Latent Superposition for Efficient Visual Reasoning
Open paper
Benchmark
Model
Rank
Results
visual-question-answering-vqa-on-blink
Laser (Qwen2.5-VL-7B-Instruct)
#34
Accuracy: 56.92
vision-language-understanding-on-mmstar
Laser (Qwen2.5-VL-7B-Instruct)
#82
Accuracy: 60.27
Rank counts only results with a code link.