List your company

Hallucination Detection on TruthfulQA AR

87.8F1-ScoreCrossHallu
31.546446.150760.75575.3593Jul 4, 2026

Evaluation results

MethodLinks
CrossHalluBackbone=Aya, Train dataset=TruthfulQA AR, Transfer setting group=Same-language baseline (BS)2026.0787.876.4690.63
CrossHalluBackbone=Phi4-mini, Train dataset=TruthfulQA AR, Transfer setting group=Same-language baseline (BS)2026.0787.0679.3492.2
CrossHalluBackbone=Phi4-mini, Train dataset=HalluScore, Transfer setting group=Cross-domain (CD)2026.0786.870.588.67
CrossHalluBackbone=Aya, Train dataset=HalluScore, Transfer setting group=Cross-domain (CD)2026.0786.872.7688.38
CrossHalluBackbone=Aya, Train dataset=TruthfulQA EN, Transfer setting group=Cross-lingual (CL)2026.0786.2574.7990.23
CrossHalluBackbone=Allam, Train dataset=TruthfulQA AR, Transfer setting group=Same-language baseline (BS)2026.0784.4365.0483.64
CrossHalluBackbone=Phi4-mini, Train dataset=TruthfulQA EN, Transfer setting group=Cross-lingual (CL)2026.0784.2579.0793.44
CrossHalluBackbone=Silma, Train dataset=TruthfulQA AR, Transfer setting group=Same-language baseline (BS)2026.0783.179.1186.17
CrossHalluBackbone=Mistral, Train dataset=TruthfulQA AR, Transfer setting group=Same-language baseline (BS)2026.0782.5380.4289.49
CrossHalluBackbone=Allam, Train dataset=HalluScore, Transfer setting group=Cross-domain (CD)2026.0781.2345.4972.42
CrossHalluBackbone=Mistral, Train dataset=HalluScore, Transfer setting group=Cross-domain (CD)2026.0780.575.4287.2
CrossHalluBackbone=Allam, Train dataset=TruthfulQA EN, Transfer setting group=Cross-lingual (CL)2026.0778.8150.2772.49
CrossHalluBackbone=Silma, Train dataset=HalluScore, Transfer setting group=Cross-domain (CD)2026.0778.6274.2580.63
CrossHalluBackbone=Silma, Train dataset=TruthfulQA EN, Transfer setting group=Cross-lingual (CL)2026.0774.5840.2860.88
CrossHalluBackbone=Mistral, Train dataset=TruthfulQA EN, Transfer setting group=Cross-lingual (CL)2026.0774.0564.2476.72
CrossHalluBackbone=Qwen2.5, Train dataset=TruthfulQA EN, Transfer setting group=Cross-lingual (CL)2026.0754.0268.5256.41
CrossHalluBackbone=Qwen2.5, Train dataset=HalluScore, Transfer setting group=Cross-domain (CD)2026.0752.5970.0657.2
CrossHalluBackbone=Qwen2.5, Train dataset=TruthfulQA AR, Transfer setting group=Same-language baseline (BS)2026.0733.7168.4651.83