List your company

Question Answering on HotpotQA (AUROC, AUPRC)

63.1AUROCP(True) + CoT-UQ
53.188855.761958.33560.9081Feb 24, 2025

Evaluation results

MethodLinks
P(True) + CoT-UQBackbone=Llama 3.1-8B, CoT-UQ=true2025.0263.137.29
P(True)Backbone=Llama 3.1-8B, CoT-UQ=false2025.0262.3935.68
Probas-mean + CoT-UQBackbone=Llama 3.1-8B, CoT-UQ=true2025.0262.0132.25
TOKENSAR + CoT-UQBackbone=Llama 3.1-8B, CoT-UQ=true2025.0261.0731.27
Probas-meanBackbone=Llama 3.1-8B, CoT-UQ=false2025.0253.7329.14
TOKENSARBackbone=Llama 3.1-8B, CoT-UQ=false2025.0253.5728.41