List your company

Object Hallucination Assessment in Image Captioning on MSCOCO 2014 (val)

59CS ScoreBeam Search
17.60828.35439.149.846May 24, 2026

Evaluation results

MethodLinks
Beam SearchModel=Shikra-7B, Decoding Strategy=Beam Search, Max new tokens=5122026.055916.374.6
VCDModel=LLaVA-1.5-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0558.618.272.8
GreedyModel=Shikra-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0557.615.775.3
VCDModel=Shikra-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0556.415.575.2
Beam SearchModel=LLaVA-1.5-7B, Decoding Strategy=Beam Search, Max new tokens=5122026.0555.615.477.5
VCDModel=LLaVA-1.5-13B, Decoding Strategy=Greedy, Max new tokens=5122026.0553.615.375.8
GreedyModel=LLaVA-1.5-7B, Decoding Strategy=Greedy, Max new tokens=5122026.055315.676.7
Beam SearchModel=LLaVA-1.5-13B, Decoding Strategy=Beam Search, Max new tokens=5122026.0550.413.879
GreedyModel=LLaVA-1.5-13B, Decoding Strategy=Greedy, Max new tokens=5122026.0549.814.678.2
OPERAModel=LLaVA-1.5-7B, Max new tokens=5122026.0545.613.179.1
OPERAModel=LLaVA-1.5-13B, Max new tokens=5122026.0542.613.277.8
OPERAModel=Shikra-7B, Max new tokens=5122026.0541.413.773.5
VCDModel=MiniGPT-4-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0541.414.168.2
PAIModel=Shikra-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0538.610.176.2
PAIModel=LLaVA-1.5-13B, Decoding Strategy=Greedy, Max new tokens=5122026.05339.277.8
GreedyModel=MiniGPT-4-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0531.81271.1
Beam SearchModel=MiniGPT-4-7B, Decoding Strategy=Beam Search, Max new tokens=5122026.0529.29.971.2
DevilsModel=LLaVA-1.5-13B, Decoding Strategy=Greedy, Max new tokens=5122026.0525.88.877.3
OPERAModel=MiniGPT-4-7B, Max new tokens=5122026.0525.49.671.2
DevilsModel=LLaVA-1.5-7B, Decoding Strategy=Greedy, Max new tokens=5122026.05256.776.1
PAIModel=LLaVA-1.5-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0524.27.175.2
DevilsModel=Shikra-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0523.89.472.7
PAIModel=MiniGPT-4-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0523.28.271.4
DevilsModel=MiniGPT-4-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0521.4870.8
SADIModel=LLaVA-1.5-13B, Decoding Strategy=Greedy, Max new tokens=5122026.0521.25.977.5
SADIModel=LLaVA-1.5-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0520.44.976.4
SADIModel=Shikra-7B, Decoding Strategy=Greedy, Max new tokens=5122026.05205.974.8
SADIModel=MiniGPT-4-7B, Decoding Strategy=Greedy, Max new tokens=5122026.0519.25.770.4