Knowledge Deficiency Hallucination Open-ended Evaluation (test) benchmark leaderboard · SOTA2 Research

Knowledge Deficiency Hallucination Open-ended Evaluation (test)

92.75BertScoreMedQwen

Evaluation Results

Method	Links
MedQwen2026.04		92.75	17.49	44.6	49.37	31.91	43.12	63
GPT-4o2026.04		91.71	14.6	33.24	47.77	28.76	40.76	77
LLaVA-NeXT 7B2026.04		90.16	13.81	38.34	44.75	22.04	33.83	202
LLaVA-Med-1.52026.04		89.86	12.98	41.3	43.52	21.37	32.27	176
LLaVA-NeXT 13B2026.04		89.56	12.03	39.46	41.59	20.16	30.7	209
LLaVA-Med2026.04		89.51	11.59	40.68	41.44	19.99	30.44	192
LLM-CXR2026.04		87.98	4.15	16.55	28.71	13.05	23.28	252
CheXagent2026.04		87.82	4.65	16.85	28.07	15.38	23.75	208
MiniGPT-42026.04		86.93	7.49	32.51	34.62	14.42	24.55	329
Med-Flamingo2026.04		84.52	5.33	22.74	26.24	10.17	21	369
XrayGPT2026.04		83.82	1.91	17.49	21.62	3.31	13.99	478
RadFM2026.04		79.66	7.99	25.51	32.63	14.22	24.14	230