List your company

Claim Verification on PrimeFacts Two Class

81Macro F1Llama-3.3-70B
37.3248.666071.34May 7, 2026

Evaluation results

MethodLinks
Llama-3.3-70BExtraction Mode=C (open)2026.058186
Qwen3-235BExtraction Mode=C (open)2026.058181
Qwen3-32BExtraction Mode=C (open)2026.057675
Llama-4-ScoutExtraction Mode=C (open)2026.057679
Llama-3.3-70BExtraction Mode=B (decontext.)2026.057478
Qwen3-8BExtraction Mode=C (open)2026.057372
Qwen3-14BExtraction Mode=C (open)2026.057277
Llama-4-ScoutExtraction Mode=B (decontext.)2026.057065
Qwen3-235BExtraction Mode=B (decontext.)2026.056970
Llama-3.3-70BExtraction Mode=A (linked)2026.056893
Qwen3-32BExtraction Mode=B (decontext.)2026.056664
Llama-4-ScoutExtraction Mode=A (linked)2026.056574
Qwen3-8BExtraction Mode=B (decontext.)2026.056364
Qwen3-235BExtraction Mode=A (linked)2026.056282
Qwen3-14BExtraction Mode=B (decontext.)2026.056165
Qwen3-32BExtraction Mode=A (linked)2026.055971
Qwen3-8BExtraction Mode=A (linked)2026.055882
Qwen3-14BExtraction Mode=A (linked)2026.055778
Baseline2026.0539—