On-Device Text Representations Robust To Misspellings via Projections
This addresses the need for privacy-preserving and low-memory on-device NLP applications by improving robustness to common misspellings, though it is incremental as it builds on existing LSH projection methods.
The paper tackled the problem of developing robust text representations for on-device natural language applications by showing that Locality-Sensitive Hashing (LSH)-based projection networks are inherently robust to misspellings, with an average accuracy drop of only 2.94% compared to 11.44% for fine-tuned BERT models under misspelling attacks.
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%.