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

VQA Hallucination on MMHal

3.87Score

GPT-4o

1.85242.37622.93.4238Jul 29, 2025
Updated 13d ago

Evaluation Results

MethodLinks
2025.07
3.8724
2025.07
3.5426
2025.07
3.2927
2025.07
3.0240
2025.07
2.8942
2025.07
2.8945
2025.07
2.8427
2025.07
2.7846
2025.07
2.746
2025.07
2.6347
2025.07
2.5446
2025.07
2.4845
2025.07
2.4850
2025.07
2.3953
2025.07
2.3257
2025.07
2.2856
2025.07
2.1961
2025.07
2.1259
2025.07
2.167
2025.07
2.0261
2025.07
1.9367