List your company
ResearchFormerly WizWand

Hallucination Detection on CNN/DM

76.14AUROCFEPoID
53.88459.66265.4471.218May 25, 2026
Updated 5mo ago

Evaluation results

MethodLinks
FEPoIDModel=Mistral-7B-Instruct-v0.3, w=7, FST=false2026.0576.14
CurvatureModel=Mistral-7B-Instruct-v0.3, w=7, FST=false2026.0573.19
IDModel=Mistral-7B-Instruct-v0.3, w=7, FST=false2026.0571.85
RGNModel=Mistral-7B-Instruct-v0.3, w=7, FST=false2026.0570.31
Val LossModel=Mistral-7B-Instruct-v0.3, w=7, FST=false2026.0569.38
RankMeModel=Mistral-7B-Instruct-v0.3, w=7, FST=false2026.0568.69
SNRModel=Mistral-7B-Instruct-v0.3, w=7, FST=false2026.0568.11
FEPoIDModel=LlaMA-3.1-8B-Instruct, w=7, FST=false2026.0559.95
CurvatureModel=LlaMA-3.1-8B-Instruct, w=7, FST=false2026.0559.22
IDModel=LlaMA-3.1-8B-Instruct, w=7, FST=false2026.0559.18
Val LossModel=LlaMA-3.1-8B-Instruct, w=7, FST=false2026.0558.59
RGNModel=LlaMA-3.1-8B-Instruct, w=7, FST=false2026.0558.21
RankMeModel=LlaMA-3.1-8B-Instruct, w=7, FST=false2026.0557.74
SNRModel=LlaMA-3.1-8B-Instruct, w=7, FST=false2026.0554.74