tigerchen52/LOVE
ACL22 paper: Imputing Out-of-Vocabulary Embeddings with LOVE Makes Language Models Robust with Little Cost
PythonMIT
Issues
- 4
typo rate problem
#13 opened by XinyeYang - 1
Can I update the vocabulary and embedding matrix with a model other than "bert base uncase"?
#12 opened by QuangTQV - 6
bert_text_classification
#11 opened by XinyeYang - 6
- 2
Train BERT Embedding
#9 opened by MingyuKim-2933 - 2
- 2
- 5
Corrupted Extrinsic dataset
#6 opened by MingyuKim-2933 - 2
How can create Hard Negative Samples?
#5 opened by MingyuKim-2933 - 2
- 6
Extrinsic task code
#2 opened by stephantul - 2
Scores
#3 opened by stephantul