List your company

Fact-checking on LIAR-RAW

77.38PrecisionKG-CRAFT
14.699230.972147.24563.5179Jan 27, 2026

Evaluation results

MethodLinks
KG-CRAFTBackbone=Llama 3.32026.0177.3870.6773.87
KG-CRAFTBackbone=Claude 3.72026.0173.9270.8672.36
KG-CRAFTBackbone=Claude 3.52026.0162.759.3860.99
Llama 3.3 70B2026.0147.2123.5231.4
DelphiAgentBackbone=gpt-4o-mini2026.0132.7922.3326.03
FactLLAMAknow2026.0132.4632.0530.44
FactLLAMA2026.0132.3231.5729.98
L-DefenseBackbone=LLAMA22026.0131.6331.7131.4
DelphiAgentBackbone=gpt-4o2026.0131.3328.3628.36
L-DefenseBackbone=ChatGPT2026.0130.5532.230.53
EExpFND2026.0129.9129.9329.58
CofCED2026.0129.4829.5528.93
Claude 3.5 Sonnet2026.0129.2527.8328.52
Claude 3.7 Sonnet2026.0128.2626.6327.42
GenFE2026.0128.0126.1626.49
ChatGPT 3.5 Turbo2026.0125.4127.3325.11
SBERT-FC2026.0124.0922.0722.19
dEFEND2026.0123.0918.5617.51
GenFE-MT2026.0118.5519.915.15
Llama 2 7B2026.0117.1117.3715.14