lanzhuzhu issues

Results 5 issues of


                                            lanzhuzhu

Different understanding about the original paper

In README, you said: “The model produces a test F1_score of 90.9 % with ~70 epochs. The results produced in the paper for the given architecture is 91.14 ” In...

Does the training set contain clean validation set for noisy label experiments?

Hi, thank you for open-sourcing this nice library for learning-to-reweight-examples, interesting work! I have a question about how the clean validation set was used. According to the last sentence of...

Which kind of BERT version is used?

Thanks for your open sourcing. This work is really solid. I have some questions about the implementation. Which kind of BERT is used in your paper? As we know BERT_LARGE...

if a negative sign missed in the equatation 2 in the original paper

I read the paper "Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank" recently and was confused with the error function of RNTN parameters, that is equation (2) in...

Are the pre-trained model and pre-training dataset available?

Thanks for your open sourcing. It is really good work. According to your paper, the pre-trained model could achieve promising performance on both the few-shot relation extraction task (FewRel1.0) and...