Our new X account is live! Follow @wizwand_team for updates

Visual Commonsense Reasoning (Q→A) on VCR (test)

61.49Q->A Accuracy (3%, 1000 SP/C)

VILLA (MAD*) [ROBERTa/CLIP-V]

Updated 4d ago

Evaluation Results

Method	Links
VILLA (MAD*) [ROBERTa/CLIP-V] 2022.04		61.49	-	43.11	78.91
VILLA (MAD*) 2022.04		60.93	-	42.95	78.83
UNITER (MAD*) 2022.04		60.88	26.78	42.23	77.05
VL-BERT (MAD*) 2022.04		58.98	26.8	40.43	77.61
VILLA (MAD*) [ViT/CLIP-T] 2022.04		58.38	40.73	-	78.59
VILLA (Baseline) 2022.04		57.01	-	34.93	78.27
CLIP-ViLp 2022.04		53.54	34.63	-	68.36
VL-BERT (MD*) 2022.04		39.06	36.78	-	55.91
Direct Finetune 2022.04		38.1	54.82	38.23	54.23
Adapters 2022.04		36.34	-	34.41	36.42
UNITER (MD*) 2022.04		35.02	39.43	-	57.64
VL-BERT (Baseline) 2022.04		34.85	23.37	30.85	53.48
UNITER (Baseline) 2022.04		33.35	24.78	31.43	54.24