language: de
datasets:
- mc4
- wikipedia
- multilexnorm
tags:
- lexical normalization
license: apache-2.0
Fine-tuned ByT5-small for MultiLexNorm (German version)
This is the official release of the fine-tuned models for the winning entry to the W-NUT 2021: Multilingual Lexical Normalization (MultiLexNorm) shared task, which evaluates lexical-normalization systems on 12 social media datasets in 11 languages.
Our system is based on ByT5, which we first pre-train on synthetic data and then fine-tune on authentic normalization data. It achieves the best performance by a wide margin in intrinsic evaluation, and also the best performance in extrinsic evaluation through dependency parsing. In addition to these fine-tuned models, we also release the source files on GitHub and an interactive demo on Google Colab.
How to use
The model was not fine-tuned in a standard sentence-to-sentence setting – instead, it was tailored to the token-to-token definition of MultiLexNorm data. Please refer to the interactive demo on Colab notebook to learn how to use these models.
How to cite
@inproceedings{wnut-ufal,
title= "{ÚFAL} at {MultiLexNorm} 2021: Improving Multilingual Lexical Normalization by Fine-tuning {ByT5}",
author = "Samuel, David and Straka, Milan",
booktitle = "Proceedings of the 7th Workshop on Noisy User-generated Text (W-NUT 2021)",
year = "2021",
publisher = "Association for Computational Linguistics",
address = "Punta Cana, Dominican Republic"
}
ByT5 - Small
ByT5 is a tokenizer-free version of Google's T5 and generally follows the architecture of MT5.
ByT5 was only pre-trained on mC4 excluding any supervised training with an average span-mask of 20 UTF-8 characters. Therefore, this model has to be fine-tuned before it is useable on a downstream task.
ByT5 works especially well on noisy text data,e.g., google/byt5-small significantly outperforms mt5-small on TweetQA.
Paper: ByT5: Towards a token-free future with pre-trained byte-to-byte models
Authors: Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel