English

Extreme Model Compression for On-device Natural Language Understanding

Computation and Language 2020-12-02 v1 Artificial Intelligence Machine Learning

Abstract

In this paper, we propose and experiment with techniques for extreme compression of neural natural language understanding (NLU) models, making them suitable for execution on resource-constrained devices. We propose a task-aware, end-to-end compression approach that performs word-embedding compression jointly with NLU task learning. We show our results on a large-scale, commercial NLU system trained on a varied set of intents with huge vocabulary sizes. Our approach outperforms a range of baselines and achieves a compression rate of 97.4% with less than 3.7% degradation in predictive performance. Our analysis indicates that the signal from the downstream task is important for effective compression with minimal degradation in performance.

Keywords

Cite

@article{arxiv.2012.00124,
  title  = {Extreme Model Compression for On-device Natural Language Understanding},
  author = {Kanthashree Mysore Sathyendra and Samridhi Choudhary and Leah Nicolich-Henkin},
  journal= {arXiv preprint arXiv:2012.00124},
  year   = {2020}
}

Comments

Long paper at COLING 2020

R2 v1 2026-06-23T20:37:16.298Z