List your company

Hallucination Detection on GPQA (test)

57.2ANLLLlama3.2
56.58860.71964.8568.981Dec 4, 2025

Evaluation results

MethodLinks
Llama3.2Model=Llama3.22025.1257.257.263.848.256.454.959.856.863.964.564.367.1
Gemma2Model=Gemma22025.1262.663.665.250.563.261.951.6—6362.56363.7
Llama3.1Model=Llama3.12025.1264.164.159.453.35958.251.663.362.464.764.466
Falcon3Model=Falcon32025.1272.562.260.250.466.365.351.664.965.666.567.567