{L}ong{T}5: {E}fficient Text-To-Text Transformer for Long Sequences

Anonymous

Paper Link: https://openreview.net/forum?id=50Ix78vyQyg

Paper Type: Long paper (up to eight pages of content + unlimited references and appendices)

Abstract:

Recent work has shown that either (1) increasing the input length or (2) increasing model size can improve the performance of Transformer-based neural models. In this paper, we present LongT5, a new model that explores the effects of scaling both the input length and model size at the same time.
Specifically, we integrate attention ideas from long-input transformers (ETC), and adopt pre-training strategies from summarization pre-training (PEGASUS) into the scalable T5 architecture. The result is a new attention mechanism we call Transient Global (TGlobal), which mimics ETC's local/global attention mechanism, but without requiring additional side-inputs. We are able to achieve state-of-the-art results on several summarization and question answering tasks, as well as outperform the original T5 models on these tasks. We have open sourced our architecture and training code, as well as our pre-trained model checkpoints.

Presentation Mode: This paper will be presented in person in Seattle

Copyright Consent Signature (type Name Or NA If Not Transferrable): Xiaoyue Guo

Copyright Consent Name And Address: Google LLC, 1600 Amphitheater Parkway, Mountain View, CA 94043