Recently, there has been a strong interest in developing natural language applications that live on personal devices such as mobile phones, watches and IoT with the objective to preserve user privacy and have low memory. Advances in Locality-Sensitive Hashing (LSH)-based projection networks have demonstrated state-of-the-art performance in various classification tasks without explicit word (or word-piece) embedding lookup tables by computing on-the-fly text representations. In this paper, we show that the projection based neural classifiers are inherently robust to misspellings and perturbations of the input text. We empirically demonstrate that the LSH projection based classifiers are more robust to common misspellings compared to BiLSTMs (with both word-piece & word-only tokenization) and fine-tuned BERT based methods. When subject to misspelling attacks, LSH projection based classifiers had a small average accuracy drop of 2.94% across multiple classifications tasks, while the fine-tuned BERT model accuracy had a significant drop of 11.44%.
Download