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

Visual Hallucination Evaluation on HallusionBench visual questions

65.8Accuracy

GPT-4V

43.23249.09154.9560.809Jan 29, 2024
Updated 3d ago

Evaluation Results

MethodLinks
2024.01
65.8
2024.01
63.9
2024.01
60.3
2024.01
57
2024.01
56.4
2024.01
56.4
2024.01
53.6
2024.01
46.7
2024.01
46.1
2024.01
44.1