List your company

Hallucination Detection on MedHallu

1AUROCB: CAA
0.4280.57650.7250.8735May 16, 2026

Evaluation results

MethodLinks
B: CAABackbone=Llama-3.3-70B2026.051————————
F: Ans. Expect.Backbone=Llama-3.3-70B2026.051————————
Ensemble A+C+E+FBackbone=Llama-3.3-70B2026.050.99————————
DRIFT-logpBackbone=Llama-3.3-70B2026.050.87————————
TXTEMB (ctrl)Backbone=Llama-3.3-70B2026.050.71————————
P(True)Backbone=Llama-3.3-70B2026.050.58————————
D: Act. VarianceBackbone=Llama-3.3-70B2026.050.57————————
DRIFT-concat (ablation of E)Backbone=Llama-3.3-70B2026.050.57————————
DoLaBackbone=Llama-3.3-70B2026.050.57————————
SAPLMABackbone=Llama-3.3-70B2026.050.57————————
E: DRIFTBackbone=Llama-3.3-70B2026.050.55————————
SelfCheckGPT-NLIBackbone=Llama-3.3-70B2026.050.55————————
A: PerturbationBackbone=Llama-3.3-70B2026.050.52————————
ACT (detect)Backbone=Llama-3.3-70B2026.050.52————————
HaMIBackbone=Llama-3.3-70B2026.050.52————————
IRIS⋆ (self-verify)Backbone=Llama-3.3-70B2026.050.51————————
HaloScopeBackbone=Llama-3.3-70B2026.050.51————————
C: Sem. EntropyBackbone=Llama-3.3-70B2026.050.5————————
INSIDEBackbone=Llama-3.3-70B2026.050.5————————
HalluShiftBackbone=Llama-3.3-70B2026.050.5————————
SEPsBackbone=Llama-3.3-70B2026.050.49————————
MINDBackbone=Llama-3.3-70B2026.050.48————————
Log-ProbBackbone=Llama-3.3-70B2026.050.48————————
PRISMBackbone=Llama-3.3-70B2026.050.45————————
EigenTrackModel=GPT-2, Evaluation=length-controlled2026.05———————0.57—
EigenTrackModel=BERT, Evaluation=length-controlled2026.05———————0.54—
EigenTrackModel=Pythia-160M, Evaluation=length-controlled2026.05———————0.61—
EigenTrackModel=Flan-T5 (dec), Evaluation=length-controlled2026.05———————0.5—
LLM-CheckModel=GPT-2, Evaluation=length-controlled2026.05————0.620.650.64——
LLM-CheckModel=BERT, Evaluation=length-controlled2026.05————0.640.72———
LLM-CheckModel=Pythia-160M, Evaluation=length-controlled2026.05————0.620.620.58——
LLM-CheckModel=Flan-T5 (cross), Evaluation=length-controlled2026.05————0.550.5———
LLM-CheckModel=Flan-T5 (dec), Evaluation=length-controlled2026.05————0.590.60.6——
StackingAnalyzer=Qwen2.5-0.5B2026.05————————0.974
StackingAnalyzer=Pythia-1.4B2026.05————————0.952
StackingAnalyzer=Gemma-2-2B2026.05————————0.965
StackingAnalyzer=LLaMA-3-3B2026.05————————0.977
StackingAnalyzer=Qwen2.5-7B2026.05————————0.99
StackingAnalyzer=LLaMA-3-8B2026.05————————0.983
StackingAnalyzer=Gemma-2-9B2026.05————————0.985
Transport DiagnosticsModel=GPT-2, Evaluation=length-controlled2026.05—0.580.610.55—————
Transport DiagnosticsModel=BERT, Evaluation=length-controlled2026.05—0.580.540.63—————
Transport DiagnosticsModel=Pythia-160M, Evaluation=length-controlled2026.05—0.610.620.59—————
Transport DiagnosticsModel=Flan-T5 (cross), Evaluation=length-controlled2026.05—0.560.670.68—————
Transport DiagnosticsModel=Flan-T5 (dec), Evaluation=length-controlled2026.05—0.530.570.56—————