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

Multimodal Medical Reasoning on VQA-RAD

80.45Accuracy (%)

Dialectic-Med

Updated 25d ago

Evaluation Results

Method	Links
Dialectic-Med 2026.04		80.45	4,500	22.5
Dialectic-Med 2026.04		79.6	4,000	16
VISTA-SFT 2026.05		78.49	-	-
Dialectic-Med 2026.04		78.24	3,800	15.2
Dialectic-Med 2026.04		76.65	4,200	22.68
SFT-Oracle 2026.05		76.49	-	-
GPT-5.1 2026.04		76.4	1,200	6
SFT-Seed 2026.05		75.7	-	-
Gemini-3 Pro 2026.04		75.15	1,100	4.4
MedMMV 2026.04		74.87	9,000	36
Dialectic-Med 2026.04		74.62	3,000	0.35
STaR 2026.05		74.5	-	-
GPT-4o 2026.04		74.28	1,000	4
MedVCD 2026.04		73.2	8,500	34
Claude-4.5 Sonnet 2026.04		72.82	1,200	6.48
VISTA-SFT 2026.05		72.51	-	-
STaR+ 2026.05		72.51	-	-
R3V 2026.05		72.51	-	-
DS-RL + GTP 2026.06		72.51	-	-
R3V 2026.05		72.31	-	-
SFT-Oracle 2026.05		72.11	-	-
GTP 2026.06		71.71	-	-
ReConcile 2026.04		71.31	12,000	48
VISTA-SFT 2026.05		71.31	-	-
DS-RL + GTP 2026.06		71.31	-	-
DS-RL + GTP 2026.06		70.52	-	-
Instruct Model 2026.06		70.52	-	-
DS-RL 2026.06		70.52	-	-
Dialectic-Med 2026.04		70.35	2,500	0.15
STaR+ 2026.05		69.32	-	-
DS-RL 2026.06		69.32	-	-
GTP 2026.06		69.32	-	-
Qwen3-VL-32B 2026.04		68.96	900	0.1
SFT-Seed 2026.05		68.92	-	-
SFT-Oracle 2026.05		68.92	-	-
STaR+ 2026.05		68.92	-	-
Med-PaLM 2 2026.04		68.65	800	1.5
DS-RL 2026.06		68.13	-	-
STaR 2026.05		66.53	-	-
Instruct Model 2026.06		66.53	-	-
R3V 2026.05		66.14	-	-
STaR 2026.05		65.34	-	-
GTP 2026.06		65.34	-	-
MedAgent 2026.04		65.33	15,000	60
Instruct Model 2026.06		64.14	-	-
Qwen3-VL-8B 2026.04		62.5	800	0.05
SFT-Seed 2026.05		58.96	-	-
LLaVA-Med 2026.04		56.2	700	0.05