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

Medical Visual Question Answering (True/False) on MIMIC (test)

73.2Accuracy

LIAR

Updated 2mo ago

Evaluation Results

Method	Links
LIAR 2026.05		73.2
Clean RAG (avg.) 2026.05		73.16
M³Att 2026.05		66.7
LIAR 2026.05		66.4
w/o RAG 2026.05		66
Clean RAG (avg.) 2026.05		65.88
M³Att 2026.05		63.46
LIAR 2026.05		61.86
w/o RAG 2026.05		61.34
LIAR 2026.05		61.04
LIAR 2026.05		58.46
M³Att 2026.05		55.44
M³Att 2026.05		51.3
M³Att 2026.05		50
Clean RAG (avg.) 2026.05		41.26
LIAR 2026.05		40.96
Clean RAG (avg.) 2026.05		40.6
Clean RAG (avg.) 2026.05		37.84
LIAR 2026.05		37.22
M³Att 2026.05		36.78
LIAR 2026.05		36.18
LIAR 2026.05		35.9
LIAR 2026.05		34.48
M³Att 2026.05		33.76
LIAR 2026.05		32.92
M³Att 2026.05		30.78
LIAR 2026.05		30.62
M³Att 2026.05		29.58
M³Att 2026.05		28.68
LIAR 2026.05		26.54
M³Att 2026.05		24.5
w/o RAG 2026.05		24
M³Att 2026.05		23.34
w/o RAG 2026.05		22.66
M³Att 2026.05		21.34
LIAR 2026.05		21.08
M³Att 2026.05		14
w/o RAG 2026.05		0
LIAR 2026.05		0
M³Att 2026.05		0