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

Claim Verification on PrimeFacts Two Class

81Macro F1

Llama-3.3-70B

Updated 26d ago

Evaluation Results

Method	Links
Llama-3.3-70B 2026.05		81	86
Qwen3-235B 2026.05		81	81
Qwen3-32B 2026.05		76	75
Llama-4-Scout 2026.05		76	79
Llama-3.3-70B 2026.05		74	78
Qwen3-8B 2026.05		73	72
Qwen3-14B 2026.05		72	77
Llama-4-Scout 2026.05		70	65
Qwen3-235B 2026.05		69	70
Llama-3.3-70B 2026.05		68	93
Qwen3-32B 2026.05		66	64
Llama-4-Scout 2026.05		65	74
Qwen3-8B 2026.05		63	64
Qwen3-235B 2026.05		62	82
Qwen3-14B 2026.05		61	65
Qwen3-32B 2026.05		59	71
Qwen3-8B 2026.05		58	82
Qwen3-14B 2026.05		57	78
Baseline 2026.05		39	-