Raphael Tang

JavaScript Convolutional Neural Networks for Keyword Spotting in the Browser: An Experimental Analysis

2018 · arXiv (Cornell University)

Used for simple commands recognition on devices from smart routers to mobile phones, keyword spotting systems are everywhere. Ubiquitous as well are web applications, which have grown in popularity and complexity over the last decade …

“Low-Resource” Text Classification: A Parameter-Free Classification Method with Compressors

2023

Deep neural networks (DNNs) are often used for text classification due to their high accuracy. However, DNNs can be computationally intensive, requiring millions of parameters and large amounts of labeled data, which can make them …

Distilling Task-Specific Knowledge from BERT into Simple Neural Networks

2019 · arXiv (Cornell University)

In the natural language processing literature, neural networks are becoming increasingly deeper and complex. The recent poster child of this trend is the deep language representation model, which includes BERT, ELMo, and GPT. These developments …

DocBERT: BERT for Document Classification

2019 · arXiv (Cornell University)

We present, to our knowledge, the first application of BERT to document classification. A few characteristics of the task might lead one to think that BERT is not the most appropriate model: syntactic structures matter …

DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference

2020

Large-scale pre-trained language models such as BERT have brought significant improvements to NLP applications. However, they are also notorious for being slow in inference, which makes them difficult to deploy in realtime applications. We propose …