SqueezeBERT: What can computer vision teach NLP about efficient neural networks?

Humans read and write hundreds of billions of messages every day. Further, due to the availability of large datasets, large computing systems, and better neural network models, natural language processing (NLP) technology has made significant strides in understanding, proofreading, and organizing these messages. Thus, there is a significant opportunity to deploy NLP in myriad applications to help web users, social networks, and businesses. In particular, we consider smartphones and other mobile devices as crucial platforms for deploying NLP models at scale. However, today's highly-accurate NLP neural network models such as BERT and RoBERTa are extremely computationally expensive, with BERT-base taking 1.7 seconds to classify a text snippet on a Pixel 3 smartphone. In this work, we observe that methods such as grouped convolutions have yielded significant speedups for computer vision networks, but many of these techniques have not been adopted by NLP neural network designers. We demonstrate how to replace several operations in self-attention layers with grouped convolutions, and we use this technique in a novel network architecture called SqueezeBERT, which runs 4.3x faster than BERT-base on the Pixel 3 while achieving competitive accuracy on the GLUE test set. The SqueezeBERT code will be released.

PDF Abstract EMNLP (sustainlp) 2020 PDF EMNLP (sustainlp) 2020 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Linguistic Acceptability CoLA SqueezeBERT Accuracy 46.5% # 32
Semantic Textual Similarity MRPC SqueezeBERT Accuracy 87.8% # 21
Number of Params 51.1M # 5
Natural Language Inference MultiNLI SqueezeBERT Matched 82.0 # 33
Mismatched 81.1 # 26
Natural Language Inference QNLI SqueezeBERT Accuracy 90.1% # 32
Question Answering Quora Question Pairs SqueezeBERT Accuracy 80.3% # 18
Natural Language Inference RTE SqueezeBERT Accuracy 73.2% # 30
Sentiment Analysis SST-2 Binary classification SqueezeBERT Accuracy 91.4 # 48
Natural Language Inference WNLI SqueezeBERT Accuracy 65.1% # 13