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

PitVQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Closed-ended Visual Question AnsweringPitVQA
F1 Score60.43
26
Binary safety hallucination detectionPitVQA (External val)
Accuracy96
25
Visual Question AnsweringOpen-ended PitVQA (External val)
BLEU0.441
5
Binary safety hallucination detectionPitVQA Open-ended (External val)
SNNE91
5
Surgical Visual Question AnsweringPitVQA
FScore59.52
4
Showing 5 of 5 rows