Self_Explaining_Structures_Improve_NLP_Models icon indicating copy to clipboard operation
Self_Explaining_Structures_Improve_NLP_Models copied to clipboard

Published 20 hours ago • verified publisher icon ShannonAI

Fail to reproduce the training result on SNLI

Open ZacharyChenpk opened this issue 2 years ago • 0 comments

I have tried the training script in this repository to train a model for SNLI dataset, and found that when training process continued, the predicted result of all valid cases seemed to fall into the same majority class, which made the accuracy fell to 0.3382. What should I do to reproduce the results in the paper?

ZacharyChenpk avatar Oct 04 '22 07:10 ZacharyChenpk