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

Multimodal Hallucination Evaluation on MMHal-Bench

4.67Average Score

Self-Aug

1.77882.52943.284.0306Oct 15, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
4.6729
2025.10
4.6329
2025.10
4.5631
2025.10
4.5232
2025.10
2.5559
2025.10
2.5360
2025.10
2.5261
2025.10
2.3764
2025.10
2.3565
2025.10
2.3265
2025.10
2.3264
2025.10
2.2765
2025.10
2.2150
2025.10
2.2150
2025.10
2.1751
2025.10
2.1664
2025.10
2.1550
2025.10
2.0368
2025.10
1.9970
2025.10
1.8969