DeltaLM: Encoder-Decoder Pre-training for Language Generation and Translation by Augmenting Pretrained Multilingual Encoders

TL;DR

DeltaLM is a novel pretrained multilingual encoder-decoder model that bridges the gap between language understanding and generation tasks by augmenting existing encoders with a decoder and employing self-supervised pre-training.

Contribution

It introduces a unified encoder-decoder pretraining framework that leverages both monolingual and bilingual data for improved language generation and translation performance.

Findings

Outperforms strong baselines on translation and generation tasks

Effective use of span corruption and translation span corruption pretraining tasks

Achieves state-of-the-art results in multiple NLP benchmarks

Abstract

While pretrained encoders have achieved success in various natural language understanding (NLU) tasks, there is a gap between these pretrained encoders and natural language generation (NLG). NLG tasks are often based on the encoder-decoder framework, where the pretrained encoders can only benefit part of it. To reduce this gap, we introduce DeltaLM, a pretrained multilingual encoder-decoder model that regards the decoder as the task layer of off-the-shelf pretrained encoders. Specifically, we augment the pretrained multilingual encoder with a decoder and pre-train it in a self-supervised way. To take advantage of both the large-scale monolingual data and bilingual data, we adopt the span corruption and translation span corruption as the pre-training tasks. Experiments show that DeltaLM outperforms various strong baselines on both natural language generation and translation tasks, including machine translation, abstractive text summarization, data-to-text, and question generation. The code and pretrained models are available at \url{https://aka.ms/deltalm}.