ReadOnce Transformers: Reusable Representations of Text for Transformers

TL;DR

ReadOnce Transformers create reusable, compressed text representations that enable faster training and evaluation, and can handle longer documents without new pre-training, benefiting multiple downstream tasks.

Contribution

The paper introduces ReadOnce Transformers, a novel method for generating reusable, task-independent text representations that improve efficiency and scalability in transformer models.

Findings

2x-5x speedup in training and evaluation

Enables handling of longer documents

Improves performance on multiple downstream tasks

Abstract

We present ReadOnce Transformers, an approach to convert a transformer-based model into one that can build an information-capturing, task-independent, and compressed representation of text. The resulting representation is reusable across different examples and tasks, thereby requiring a document shared across many examples or tasks to only be \emph{read once}. This leads to faster training and evaluation of models. Additionally, we extend standard text-to-text transformer models to Representation+Text-to-text models, and evaluate on multiple downstream tasks: multi-hop QA, abstractive QA, and long-document summarization. Our one-time computed representation results in a 2x-5x speedup compared to standard text-to-text models, while the compression also allows existing language models to handle longer documents without the need for designing new pre-trained models.