Compression of Deep Learning Models for Text: A Survey

TL;DR

This survey reviews various methods for compressing large deep learning models in NLP, such as pruning and quantization, to facilitate their deployment in real-world applications with limited resources.

Contribution

It systematically categorizes and summarizes recent advances in NLP model compression techniques, providing a coherent overview for researchers and practitioners.

Findings

Six main compression methods identified and explained.

Comprehensive organization of recent NLP model compression research.

Highlights the importance of model efficiency for industry deployment.

Abstract

In recent years, the fields of natural language processing (NLP) and information retrieval (IR) have made tremendous progress thanksto deep learning models like Recurrent Neural Networks (RNNs), Gated Recurrent Units (GRUs) and Long Short-Term Memory (LSTMs)networks, and Transformer [120] based models like Bidirectional Encoder Representations from Transformers (BERT) [24], GenerativePre-training Transformer (GPT-2) [94], Multi-task Deep Neural Network (MT-DNN) [73], Extra-Long Network (XLNet) [134], Text-to-text transfer transformer (T5) [95], T-NLG [98] and GShard [63]. But these models are humongous in size. On the other hand,real world applications demand small model size, low response times and low computational power wattage. In this survey, wediscuss six different types of methods (Pruning, Quantization, Knowledge Distillation, Parameter Sharing, Tensor Decomposition, andSub-quadratic Transformer based methods) for compression of such models to enable their deployment in real industry NLP projects.Given the critical need of building applications with efficient and small models, and the large amount of recently published work inthis area, we believe that this survey organizes the plethora of work done by the 'deep learning for NLP' community in the past fewyears and presents it as a coherent story.