Medical Visual Question Answering on Med-HallMark benchmark leaderboard · SOTA2 Research

Medical Visual Question Answering on Med-HallMark

55.11BertScoremPLUG-Owl2

Evaluation Results

Method	Links
mPLUG-Owl22024.06		55.11	29.39	22.25	8.38	19.77	3.43	0.5	24
LLaVA1.5-7b2024.06		54.89	28.33	23.52	9.3	21.16	4.6	0.53	28
LLaVA1.5-13b2024.06		52.82	25.98	21.52	8.2	19.38	4.18	0.51	23
BLIP22024.06		47.97	16.15	18.98	6.03	17.13	3.46	0.52	27
XrayGPT2024.06		44.4	14	10.66	1.17	9.89	0.37	0.36	2
RadFM2024.06		43.84	11.81	11.31	2.16	10.68	1.55	0.56	32
mini-gpt42024.06		42.93	12.93	11.14	1.6	10.19	0.39	0.38	9
LLaVA-Med2024.06		36.67	8.8	9.17	1.4	9.1	0.03	0.59	35
InstructBLIP13b2024.06		36.02	7.59	6.13	0.58	5.32	1	0.58	34
InstructBLIP2024.06		35.99	7.47	6.08	0.59	5.3	1.03	0.57	34
LLaVA-Med2024.06		35.25	6.91	6.34	1.49	6.09	0.54	0.57	32
LLaVA-Med2024.06		33.32	3.27	2.85	0.58	2.68	0.06	0.59	35