List your company

Math Word Problem Reasoning on SVAMP (AUROC)

65.83AUROC ScoreProbas-mean + CoT-UQ
42.01448.19754.3860.563Feb 24, 2025

Evaluation results

MethodLinks
Probas-mean + CoT-UQStrategy=AP, Backbone=Mistral 7B2025.0265.83
TOKENSAR + CoT-UQStrategy=AP, Backbone=Mistral 7B2025.0265.42
TOKENSARStrategy=AP, Backbone=Mistral 7B2025.0260.69
Probas-meanStrategy=AP, Backbone=Mistral 7B2025.0259.88
P(True) + CoT-UQStrategy=SE, Backbone=Mistral 7B2025.0250.37
P(True)Strategy=SE, Backbone=Mistral 7B2025.0242.93