List your company

Fact-checking on FEVER (F1Mac, AUC, Acc, ∆Acc)

94.3F1 MacroNone (Original Claims)
0.59624.92349.2573.577Jan 23, 2026

Evaluation results

MethodLinks
None (Original Claims)Input Setting=Gold-Evidence2026.0194.397.994.3—
ParaphrasingInput Setting=Gold-Evidence2026.0191.295.391.2-0.031
Word SwapInput Setting=Gold-Evidence2026.0189.795.789.7-0.046
Character PerturbationsInput Setting=Gold-Evidence2026.0186.893.886.9-0.074
Synonym SubstitutionInput Setting=Gold-Evidence2026.0184.392.784.5-0.098
None (Original Claims)Input Setting=Claim-Only2026.0181.489.381.4—
Synonym SubstitutionInput Setting=Claim-Only2026.0176.484.876.4-0.05
Persuasion (Blind)Input Setting=Gold-Evidence2026.0176.486.276.4-0.179
Character PerturbationsInput Setting=Claim-Only2026.0176.284.676.2-0.052
Word SwapInput Setting=Claim-Only2026.0175.584.775.7-0.057
ParaphrasingInput Setting=Claim-Only2026.0175.583.775.7-0.057
Persuasion (Blind)Input Setting=Claim-Only2026.0167.874.568.2-0.132
Persuasion (Oracle)Input Setting=Gold-Evidence2026.0114.22.916.4-0.779
Persuasion (Oracle)Input Setting=Claim-Only2026.014.20.44.3-0.771