EdgeFormer: A Parameter-Efficient Transformer for On-Device Seq2seq Generation

TL;DR

EdgeFormer is a novel, parameter-efficient Transformer designed for on-device sequence-to-sequence generation, outperforming previous models under strict resource constraints and enabling practical on-device NLP applications.

Contribution

It introduces two new principles for cost-effective parameterization and a layer adaptation technique, advancing on-device Transformer performance with shared layers.

Findings

Outperforms previous parameter-efficient Transformers

Achieves competitive results under resource constraints

First publicly available pretrained on-device seq2seq model

Abstract

We introduce EdgeFormer -- a parameter-efficient Transformer for on-device seq2seq generation under the strict computation and memory constraints. Compared with the previous parameter-efficient Transformers, EdgeFormer applies two novel principles for cost-effective parameterization, allowing it to perform better given the same parameter budget; moreover, EdgeFormer is further enhanced by layer adaptation innovation that is proposed for improving the network with shared layers. Extensive experiments show EdgeFormer can effectively outperform previous parameter-efficient Transformer baselines and achieve competitive results under both the computation and memory constraints. Given the promising results, we release EdgeLM -- the pretrained version of EdgeFormer, which is the first publicly available pretrained on-device seq2seq model that can be easily fine-tuned for seq2seq tasks with strong results, facilitating on-device seq2seq generation in practice.