Multi-modal Multiple Choice Question Answering on PathMMU (test) benchmark leaderboard · SOTA2 Research

Multi-modal Multiple Choice Question Answering on PathMMU (test)

75.34Accuracy (Atlas)Patho-R1-7B

Evaluation Results

Method	Links
Patho-R1-7B2025.05		75.34	63.37	66.43	45.4	66.06	67.93
PathGen-LLaVA-13B2025.05		63.83	58.79	58.47	56.19	59.24	58.47
Patho-R1-3B2025.05		62.22	57.93	62.51	34.99	61.07	64.07
HuatuoGPT-Vision-7B2025.05		58.07	54.59	54.72	36.64	61.36	59.37
InternVL3-8B2025.05		54.07	50.38	50.8	39.09	54.04	53.32
Llama-3.2V-11B-cot2025.05		51.81	44.23	45.45	30.76	48.15	46.1
InternVL2.5-8B2025.05		50.06	46.98	50.62	32.84	50.02	50.87
Qwen2.5VL-3B2025.05		45.06	42.1	45.28	23.65	46.79	48.1
InternVL2-8B2025.05		43.68	40.68	44.86	23.77	44.56	45.4
Quilt-LLaVA-v1.5-7B2025.05		41.43	32.02	36.72	14.71	34.8	35.29
Qwen2.5VL-7B2025.05		41.18	38.67	43.2	24.82	42.77	39.67
Llama-3.2-11B-Vision-Instruct2025.05		41.05	36.5	37.49	26.72	38.82	39.21
Llama3.2-3B-vl2025.05		35.79	23.39	25.01	6.99	25.37	28.91
DeepSeek-VL2-Tiny2025.05		32.17	31.36	35.41	19.18	35.38	32.13
VILA1.5-3B2025.05		26.53	26.74	30.6	15.07	29.1	30.71
LLaVA-Onevision-7B2025.05		21.65	23.4	21.27	12.01	27.77	21.25
LLaVA-Med-v1.5-Mistral-7B2025.05		21.4	18.59	21.51	7.72	21.31	20.54
Paligemma2-3b-pt-2242025.05		0.5	1.64	3.15	0.06	2.01	1.61