Char2Subword: Extending the Subword Embedding Space Using Robust Character Compositionality
Gustavo Aguilar, Bryan McCann, Tong Niu, Nazneen Fatema Rajani, Nitish Shirish Keskar, Thamar Solorio · 2021
Byte-pair encoding (BPE) is a ubiquitous algorithm in the subword tokenization process of language models as it provides multiple benefits.However, this process is solely based on pre-training data statistics, making it hard for the tokenizer to handle infrequent spellings.On the other hand, though robust to misspellings, pure character-level models often lead to unreasonably long sequences and make it harder for the model to learn meaningful words.To alleviate these challenges, we propose a character-based subword module (char2subword) 1 that learns the subword embedding table in pre-trained models like BERT.Our char2subword module builds representations from characters out of the subword vocabulary, and it can be used as a dropin replacement of the subword embedding table.The module is robust to character-level alterations such as misspellings, word inflection, casing, and punctuation.We integrate it further with BERT through pre-training while keeping BERT transformer parameters fixedand thus, providing a practical method.Finally, we show that incorporating our module to mBERT significantly improves the performance on the social media linguistic codeswitching evaluation (LinCE) benchmark.* Work performed as summer intern at Salesforce.