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

Scientific Claim Verification on Matter-of-Fact scientific claims

73Overall Accuracy

GPT-5-mini + Analytica-L

Updated 2mo ago

Evaluation Results

Method	Links
GPT-5-mini + Analytica-L 2026.04		73	70	75	75	77	69	67	79	67
GPT-5-mini 2026.04		71	60	82	69	76	66	64	75	72
GPT-5.1 + Analytica-L 2026.04		70	85	55	73	75	71	67	70	64
GPT-4o-mini 2026.04		66	90	42	72	72	68	63	61	58
O4-mini + Analytica-L 2026.04		64	95	30	57	60	73	62	68	66
GPT-5.1 2026.04		62	27	97	58	56	59	69	66	62
O4-mini 2026.04		61	34	88	60	57	63	62	59	63
GPT-4o-mini + Analytica-L 2026.04		59	87	30	55	58	59	59	60	60
Random 2026.04		50	50	50	50	50	50	50	50	50