Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multi-modal Hallucination Evaluation on MMHal-Bench

1.83Attribute Failure Score

PAI

1.82.00252.2052.4075Jun 28, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2026.06
1.830.752.3321.922.171.252.421.83
2026.06
1.830.831.831.171.674.171.582.251.92
2026.06
2.251.332.921.751.923.251.581.422.05
2026.06
2.421.252.831.752.083.251.421.752.09
2026.06
2.581.3331.751.923.421.171.422.07