Journal Article

ByT5: Towards a Token-Free Future with Pre-trained Byte-to-Byte Models

Xue, Linting, Raffel, Colin, Constant, Noah, Al-Rfou, Rami, Kale, Mihir, Narang, Sharan, Roberts, Adam, Barua, Aditya

2022

Overview

Most widely used pre-trained language models operate on sequences of tokens corresponding to word or subword units. By comparison, models that operate directly on raw text (bytes or characters) have many benefits: They can process text in any language out of the box, they are more robust to noise, and they minimize technical debt by removing complex and error-prone text preprocessing pipelines. Because byte or character sequences are longer than token sequences, past work on token-free models has often introduced new model architectures designed to amortize the cost of operating directly on raw text. In this paper, we show that a standard Transformer architecture can be used with minimal modifications to process byte sequences. We characterize the trade-offs in terms of parameter count, training FLOPs, and inference speed, and show that byte-level models are competitive with their token-level counterparts. We also demonstrate that byte-level models are significantly more robust to noise and perform better on tasks that are sensitive to spelling and pronunciation. As part of our contribution, we release a new set of pre-trained byte-level Transformer models based on the T5 architecture, as well as all code and data used in our experiments.

• Publisher: MIT Press,MIT Press Journals, The,The MIT Press
• Subject: Amortization / Computational linguistics / Experiments / Inference / Language / Language modeling / Linguistics / Multilingualism / Noise / Pipelines / Pronunciation / Robustness / Sequences / Spelling / Subword units