List your company
ResearchFormerly WizWand

Hallucination Detection on HotpotQA (Accuracy)

80.3AccuracyVal Loss
69.681672.438375.19577.9517May 25, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Val LossModel=LlaMA-3.2-1B, Forward horizon (w)=3, Representation extraction (FST)=true2026.0580.3
FEPoIDModel=LlaMA-3.2-1B, Forward horizon (w)=3, Representation extraction (FST)=true2026.0580.3
FEPoIDModel=LlaMA-3.2-3B, Forward horizon (w)=7, Representation extraction (FST)=true2026.0579.92
CurvatureModel=LlaMA-3.2-1B, Forward horizon (w)=3, Representation extraction (FST)=true2026.0579.57
Val LossModel=LlaMA-3.2-3B, Forward horizon (w)=7, Representation extraction (FST)=true2026.0579
CurvatureModel=LlaMA-3.2-3B, Forward horizon (w)=7, Representation extraction (FST)=true2026.0577.79
SNRModel=LlaMA-3.2-3B, Forward horizon (w)=7, Representation extraction (FST)=true2026.0577.79
SNRModel=LlaMA-3.2-1B, Forward horizon (w)=3, Representation extraction (FST)=true2026.0576.94
RGNModel=LlaMA-3.2-3B, Forward horizon (w)=7, Representation extraction (FST)=true2026.0574.93
RGNModel=LlaMA-3.2-1B, Forward horizon (w)=3, Representation extraction (FST)=true2026.0574.85
IDModel=LlaMA-3.2-1B, Forward horizon (w)=3, Representation extraction (FST)=true2026.0574.85
IDModel=LlaMA-3.2-3B, Forward horizon (w)=7, Representation extraction (FST)=true2026.0573.09
RankMEModel=LlaMA-3.2-3B, Forward horizon (w)=7, Representation extraction (FST)=true2026.0570.83
RankMEModel=LlaMA-3.2-1B, Forward horizon (w)=3, Representation extraction (FST)=true2026.0570.09