ByT5: Towards a token-free future with pre-trained byte-to-byte models

ByT5 is a tokenizer-free extension of the mT5 model. Instead of using a subword vocabulary like most other pretrained language models (BERT, XLM-R, T5, GPT-3), our ByT5 model operates directly on UTF-8 bytes, removing the need for any text preprocessing. Beyond the reduction in system complexity, we find that parameter-matched ByT5 models are competitive with mT5 across a range of tasks, and outperform mT5 on tasks that involve noisy text or are sensitive to spelling and pronunciation. This repo can be used to reproduce the experiments in the ByT5 paper.

Usage

Training

To run this code, you need to install the t5 library. General instructions for training, fine-tuning, evaluation, and exporting models for inference can be found in the t5 repo. In order to use the additional ByT5 tasks provided in this library with the t5_mesh_transformer command, run from this directory and add the flag --module_import="byt5.tasks".

To train a ByT5-Large model on the mc4 task from scratch as described in the paper:

export PROJECT=yourproject
export ZONE=yourzone
export BUCKET=yourbucket
export TPU=yourtpu

ctpu up --name=$TPU --project=$PROJECT --zone=$ZONE --tpu-size=v3-256 --tpu-only --noconf

TASK=byt5_mc4
MODEL_DIR="${BUCKET}${TASK}"

python -m t5.models.mesh_transformer_main \
  --tpu="${TPU}" \
  --gcp_project="${PROJECT}" \
  --tpu_zone="${ZONE}" \
  --model_dir="${MODEL_DIR}" \
  --gin_file="models/byt5.large.gin" \
  --gin_param="MIXTURE_NAME = '${TASK}'" \
  --gin_param="utils.run.sequence_length = {'inputs': 1024, 'targets': 189}" \
  --gin_param="utils.run.batch_size = ('tokens_per_batch', 1048576)" \
  --gin_param="[email protected]_rate_schedules.rsqrt_no_ramp_down" \
  --gin_param="run.train_steps = 1000000" \
  --gin_param="utils.tpu_mesh_shape.model_parallelism = 1" \
  --gin_param="utils.tpu_mesh_shape.tpu_topology = 'v3-256'" \
  --eval_mode="perplexity_eval" \
  --eval_gin_param="mesh_eval_dataset_fn.num_eval_examples = 10000" \
  --t5_tfds_data_dir="${BUCKET}/t5-tfds" \
  --module_import="byt5.tasks"

Fine-Tuning

The example below shows how to finetune the ByT5-Large model on the XNLI zeroshot task.

export PROJECT=yourproject
export ZONE=yourzone
export BUCKET=yourbucket
export TPU=yourtpu

ctpu up --name=$TPU --project=$PROJECT --zone=$ZONE --tpu-size=v3-256 --tpu-only --noconf

TASK=byt5_xnli_zeroshot
PRETRAINED_DIR=gs://t5-data/pretrained_models/byt5/large
PRETRAINED_STEPS=1000000
FINETUNE_STEPS=262144
MODEL_DIR="${BUCKET}${TASK}"

# Run fine-tuning
python -m t5.models.mesh_transformer_main \
  --tpu="${TPU}" \
  --gcp_project="${PROJECT}" \
  --tpu_zone="${ZONE}" \
  --model_dir="${MODEL_DIR}" \
  --gin_file="${PRETRAINED_DIR}/operative_config.gin" \
  --gin_param="utils.tpu_mesh_shape.tpu_topology = 'v3-256'" \
  --gin_param="MIXTURE_NAME = '${TASK}'" \
  --gin_param="utils.run.train_steps=$((PRETRAINED_STEPS+FINETUNE_STEPS))" \
  --gin_param="utils.run.init_checkpoint='${PRETRAINED_DIR}/model.ckpt-${PRETRAINED_STEPS}'" \
  --t5_tfds_data_dir="${BUCKET}/t5-tfds" \
  --module_import="byt5.tasks"
  --gin_param="utils.run.batch_size = ('tokens_per_batch', 1048576)" \
  --gin_param="utils.run.sequence_length = {'inputs': 2048, 'targets': 56}"
  --eval_gin_param="Bitransformer.decode.max_decode_length = 56" \

The remaining experiments are shown in the tasks.py file.

Released Model Checkpoints

We have released the following checkpoints for pre-trained models described in our paper:

ByT5-Small (300 million parameters): gs://t5-data/pretrained_models/byt5/small
ByT5-Base (580 million parameters): gs://t5-data/pretrained_models/byt5/base
ByT5-Large (1.2 billion parameters): gs://t5-data/pretrained_models/byt5/large
ByT5-XL (3.7 billion parameters): gs://t5-data/pretrained_models/byt5/xl
ByT5-XXL (13 billion parameters): gs://t5-data/pretrained_models/byt5/xxl

How to Cite

If you extend or use this work, please cite the paper where it was introduced:

@misc{xue2021byt5,
    title={ByT5: Towards a token-free future with pre-trained byte-to-byte models},
    author={Linting Xue and Aditya Barua and Noah Constant and Rami Al-Rfou and Sharan Narang and Mihir Kale and Adam Roberts and Colin Raffel},
    year={2021},
    eprint={2105.13626},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

This is not an officially supported Google product.

ByT5: Towards a token-free future with pre-trained byte-to-byte models

Related tags

Overview

ByT5: Towards a token-free future with pre-trained byte-to-byte models

Usage

Training

Fine-Tuning

Released Model Checkpoints

How to Cite

Owner

Google Research

🧪 Cutting-edge experimental spaCy components and features

Header-only C++ HNSW implementation with python bindings

Predict an emoji that is associated with a text

In this project, we aim to achieve the task of predicting emojis from tweets. We aim to investigate the relationship between words and emojis.

Codes for coreference-aware machine reading comprehension

Tools for curating biomedical training data for large-scale language modeling

text to speech toolkit. 好用的中文语音合成工具箱，包含语音编码器、语音合成器、声码器和可视化模块。

Chinese Grammatical Error Diagnosis

Contact Extraction with Question Answering.

Python bindings to the dutch NLP tool Frog (pos tagger, lemmatiser, NER tagger, morphological analysis, shallow parser, dependency parser)

A natural language modeling framework based on PyTorch

中文生成式预训练模型

Pervasive Attention: 2D Convolutional Networks for Sequence-to-Sequence Prediction

An open-source NLP library: fast text cleaning and preprocessing.

Fuzzy String Matching in Python

Practical Machine Learning with Python

A pytorch implementation of the ACL2019 paper "Simple and Effective Text Matching with Richer Alignment Features".

A fast and lightweight python-based CTC beam search decoder for speech recognition.

PyTorch Implementation of "Bridging Pre-trained Language Models and Hand-crafted Features for Unsupervised POS Tagging" (Findings of ACL 2022)

Making text a first-class citizen in TensorFlow.