Our new X account is live! Follow @wizwand_team for updates

Medical Question Answering on Medical QA Benchmarks (MedQA, MedMCQA, MMLU, CMB, CMExam, CMMLU) (test)

64.1MedQA Accuracy

KaFT

Updated 3d ago

Evaluation Results

Method	Links
KaFT 2025.05		64.1	56.94	74.01	47.39	47.89	45.44	55.96	1.8
No-conflict 2025.05		61.98	56.11	73.4	47.17	47.72	45.59	55.33	1.17
Vanilla SFT 2025.05		61.82	55.75	73.34	45.99	45.85	44.95	54.62	0.46
Self-aligning 2025.05		61.35	56.56	73	47.53	48.34	46.55	55.55	1.39
Vanilla SFT 2025.05		59.86	43.75	68.06	36.25	36.25	35.48	46.61	3.33
Base 2025.05		59.62	56.51	72.66	45.5	46.04	44.62	54.16	-
KaFT 2025.05		59.54	49.87	68.47	38.11	37.35	40.73	49.01	5.73
No-conflict 2025.05		58.37	51.11	68.69	38.09	36.92	36.17	48.22	4.94
Self-aligning 2025.05		55.62	50.99	68.81	36.7	36.81	35.13	47.34	4.06
Vanilla SFT 2025.05		54.99	50.08	62.94	38	39.17	38.13	47.22	1.91
KaFT 2025.05		54.52	50.51	64.93	40.19	39.93	39.7	48.3	2.99
KaFT 2025.05		53.57	49.82	63.23	75.57	77.27	72.07	65.25	1.32
No-conflict 2025.05		52.95	49.2	63.76	38.91	39.78	37.27	46.98	1.67
Self-aligning 2025.05		52.79	49.82	62.93	38.53	38.66	38.55	46.88	1.57
Vanilla SFT 2025.05		52.71	50.2	61.6	74.3	76.15	70.16	64.19	0.26
No-conflict 2025.05		52.24	50.54	61.55	75.04	76.87	70.77	64.5	0.57
Self-aligning 2025.05		51.77	50.11	63.18	75.23	77.05	70.71	64.67	0.74
Base 2025.05		51.14	49.41	62.34	35.98	36.48	36.5	45.31	-
Base 2025.05		50.98	48.31	65.07	31.73	30.68	32.9	43.28	-
Base 2025.05		48.94	50.08	62.79	74.77	76.59	70.42	63.93	-