List your company

Visual Question Answering on ViTextVQA (test)

53.95F1 ScoreViTextBLIP-2
24.320432.012739.70547.3973Apr 16, 2024

Evaluation results

MethodLinks
ViTextBLIP-2Version=base, Method Type=ViT-LM, Model Size=≈2.5B, % Param Tuning=≈5%2024.0453.9525.48
PreSTUVersion=large, Method Type=ViT-LM, Model Size=≈860M, % Param Tuning=≈100%2024.0444.9322.64
PreSTUVersion=base, Method Type=ViT-LM, Model Size=≈315M, % Param Tuning=≈100%2024.0443.8120.85
SaLVersion=base, Method Type=CNN-LM, Model Size=≈270M, % Param Tuning=≈100%2024.0443.4920.11
SaLVersion=large, Method Type=CNN-LM, Model Size=≈780M, % Param Tuning=≈100%2024.0443.3921.48
LaTrVersion=large, Method Type=ViT-LM, Model Size=≈900M, % Param Tuning=≈100%2024.0442.3420.94
LaTrVersion=base, Method Type=ViT-LM, Model Size=≈330M, % Param Tuning=≈100%2024.0440.4117.85
BLIP-2Version=large, Method Type=ViT-LM, Model Size=≈1.4B, % Param Tuning=≈8%2024.0438.4715.58
BLIP-2Version=base, Method Type=ViT-LM, Model Size=≈450M, % Param Tuning=≈25%2024.0436.8713.59
M4CVersion=base, Method Type=CNN-LM, Model Size=≈140M, % Param Tuning=≈100%2024.0425.4610.26