List your company

Uncertainty Quantification on OKVQA, VizWiz, VQA-RAD, and AdVQA (Average)

0.691AUROCVIG-TUQ
0.544360.582430.62050.65857May 26, 2026

Evaluation results

MethodLinks
VIG-TUQBackbone=Qwen2.5-VL-3B, UQ Level=Token-level (TL)2026.050.691111
VIG-TUQ-ABackbone=Qwen2.5-VL-3B, UQ Level=Token-level (TL)2026.050.689113
KLEBackbone=llava-1.5-7B, UQ Level=Semantic-level (SL)2026.050.686121
KLEBackbone=llava-1.5-13B, UQ Level=Semantic-level (SL)2026.050.683127
KLEBackbone=idefics-9B, UQ Level=Semantic-level (SL)2026.050.682120
SEBackbone=llava-1.5-7B, UQ Level=Semantic-level (SL)2026.050.674121
SEBackbone=idefics-9B, UQ Level=Semantic-level (SL)2026.050.674121
KLEBackbone=Qwen2.5-VL-3B, UQ Level=Semantic-level (SL)2026.050.673113
VIG-TUQBackbone=llava-1.5-13B, UQ Level=Token-level (TL)2026.050.671133
SEBackbone=llava-1.5-13B, UQ Level=Semantic-level (SL)2026.050.668127
KLEBackbone=idefics2-8B, UQ Level=Semantic-level (SL)2026.050.668124
VIG-TUQBackbone=llava-1.5-7B, UQ Level=Token-level (TL)2026.050.663124
SEBackbone=Qwen2.5-VL-3B, UQ Level=Semantic-level (SL)2026.050.662114
VIG-TUQBackbone=idefics2-8B, UQ Level=Token-level (TL)2026.050.661140
VIG-TUQ-ABackbone=llava-1.5-13B, UQ Level=Token-level (TL)2026.050.66134
VIG-TUQ-ABackbone=idefics-9B, UQ Level=Token-level (TL)2026.050.66142
VIG-TUQBackbone=idefics-9B, UQ Level=Token-level (TL)2026.050.66142
CCPBackbone=idefics-9B, UQ Level=Token-level (TL)2026.050.659118
KLEBackbone=Qwen2.5-VL-7B, UQ Level=Semantic-level (SL)2026.050.658147
Max-ProbBackbone=idefics-9B, UQ Level=Token-level (TL)2026.050.656164
VIG-TUQ-JSDBackbone=llava-1.5-13B, UQ Level=Token-level (TL)2026.050.656135
SEBackbone=idefics2-8B, UQ Level=Semantic-level (SL)2026.050.655125
VIG-TUQBackbone=Qwen2.5-VL-7B, UQ Level=Token-level (TL)2026.050.654152
Log-PerplexityBackbone=Qwen2.5-VL-3B, UQ Level=Token-level (TL)2026.050.653128
VIG-TUQ-JSDBackbone=Qwen2.5-VL-3B, UQ Level=Token-level (TL)2026.050.652124
VIG-TUQ-JSDBackbone=llava-1.5-7B, UQ Level=Token-level (TL)2026.050.652126
Token-EntropyBackbone=Qwen2.5-VL-3B, UQ Level=Token-level (TL)2026.050.651126
VIG-TUQ-ABackbone=Qwen2.5-VL-7B, UQ Level=Token-level (TL)2026.050.651152
SEBackbone=Qwen2.5-VL-7B, UQ Level=Semantic-level (SL)2026.050.651148
Token-EntropyBackbone=llava-1.5-13B, UQ Level=Token-level (TL)2026.050.65140
VIG-TUQ-JSDBackbone=idefics2-8B, UQ Level=Token-level (TL)2026.050.647145
Log-PerplexityBackbone=llava-1.5-13B, UQ Level=Token-level (TL)2026.050.646139
VIG-TUQ-ABackbone=idefics2-8B, UQ Level=Token-level (TL)2026.050.645145
CCPBackbone=Qwen2.5-VL-3B, UQ Level=Token-level (TL)2026.050.644120
Max-ProbBackbone=Qwen2.5-VL-3B, UQ Level=Token-level (TL)2026.050.643150
VIG-TUQ-ABackbone=llava-1.5-7B, UQ Level=Token-level (TL)2026.050.64126
Token-SarBackbone=llava-1.5-13B, UQ Level=Token-level (TL)2026.050.639135
VIG-TUQ-ABackbone=Qwen2-1.5B, UQ Level=Token-level (TL)2026.050.633103
Token-EntropyBackbone=llava-1.5-7B, UQ Level=Token-level (TL)2026.050.632133
VIG-TUQBackbone=Qwen2-1.5B, UQ Level=Token-level (TL)2026.050.632104
Log-PerplexityBackbone=llava-1.5-7B, UQ Level=Token-level (TL)2026.050.63135
Token-EntropyBackbone=idefics-9B, UQ Level=Token-level (TL)2026.050.629154
CCPBackbone=llava-1.5-13B, UQ Level=Token-level (TL)2026.050.627137
CCPBackbone=idefics2-8B, UQ Level=Token-level (TL)2026.050.627137
Token-EntropyBackbone=idefics2-8B, UQ Level=Token-level (TL)2026.050.627146
KLEBackbone=Qwen2-1.5B, UQ Level=Semantic-level (SL)2026.050.627127
Log-PerplexityBackbone=idefics-9B, UQ Level=Token-level (TL)2026.050.625157
Max-ProbBackbone=llava-1.5-13B, UQ Level=Token-level (TL)2026.050.625163
Log-PerplexityBackbone=idefics2-8B, UQ Level=Token-level (TL)2026.050.624148
CCPBackbone=llava-1.5-7B, UQ Level=Token-level (TL)2026.050.624120
Max-ProbBackbone=llava-1.5-7B, UQ Level=Token-level (TL)2026.050.623159
Log-PerplexityBackbone=Qwen2.5-VL-7B, UQ Level=Token-level (TL)2026.050.622164
Token-EntropyBackbone=Qwen2.5-VL-7B, UQ Level=Token-level (TL)2026.050.622164
CCPBackbone=Qwen2.5-VL-7B, UQ Level=Token-level (TL)2026.050.62148
Max-ProbBackbone=idefics2-8B, UQ Level=Token-level (TL)2026.050.617162
Max-ProbBackbone=Qwen2.5-VL-7B, UQ Level=Token-level (TL)2026.050.616172
Token-EntropyBackbone=Qwen2-1.5B, UQ Level=Token-level (TL)2026.050.616109
Log-PerplexityBackbone=Qwen2-1.5B, UQ Level=Token-level (TL)2026.050.615111
VIG-TUQ-JSDBackbone=Qwen2-1.5B, UQ Level=Token-level (TL)2026.050.614108
Max-ProbBackbone=Qwen2-1.5B, UQ Level=Token-level (TL)2026.050.611112
VIG-TUQ-JSDBackbone=Qwen2.5-VL-7B, UQ Level=Token-level (TL)2026.050.611160
Token-SarBackbone=llava-1.5-7B, UQ Level=Token-level (TL)2026.050.609136
CCPBackbone=Qwen2-1.5B, UQ Level=Token-level (TL)2026.050.60499
Token-SarBackbone=idefics2-8B, UQ Level=Token-level (TL)2026.050.599153
Token-SarBackbone=Qwen2.5-VL-7B, UQ Level=Token-level (TL)2026.050.588183
Token-SarBackbone=Qwen2-1.5B, UQ Level=Token-level (TL)2026.050.587130
VIG-TUQ-JSDBackbone=idefics-9B, UQ Level=Token-level (TL)2026.050.575167
SEBackbone=Qwen2-1.5B, UQ Level=Semantic-level (SL)2026.050.555123
Token-SarBackbone=Qwen2.5-VL-3B, UQ Level=Token-level (TL)2026.050.554175
Token-SarBackbone=idefics-9B, UQ Level=Token-level (TL)2026.050.55170