List your company

Visual Question Answering on GQA (Acc., Relative Performance)

61.9AccuracyLLaVA-1.5 Vanilla 7B
45.46849.7345458.266Aug 25, 2025

Evaluation results

MethodLinks
LLaVA-1.5 Vanilla 7BVisual Tokens=5762025.0861.9100
VisionZip ๐Ÿ”ฅVisual Tokens=1922025.0860.197.1
MMTokVisual Tokens=1922025.0860.0797
DivPruneVisual Tokens=1922025.0859.9796.9
VisionZipVisual Tokens=1922025.0859.395.8
MMTokVisual Tokens=1282025.0859.2995.8
DivPruneVisual Tokens=1282025.0859.2595.7
VisionZip ๐Ÿ”ฅVisual Tokens=1282025.0858.995.2
MMTokVisual Tokens=642025.0858.2994.2
DivPruneVisual Tokens=642025.0857.7893.3
SparseVLMVisual Tokens=1922025.0857.693.1
VisionZipVisual Tokens=1282025.0857.693.1
VisionZip ๐Ÿ”ฅVisual Tokens=642025.085792.1
SparseVLMVisual Tokens=1282025.085690.5
VisionZipVisual Tokens=642025.0855.189
FastVVisual Tokens=1922025.0852.785.1
SparseVLMVisual Tokens=642025.0852.785.1
FastVVisual Tokens=1282025.0849.680.1
FastVVisual Tokens=642025.0846.174.5