PTT5: Pretraining and validating the T5 model on Brazilian Portuguese data
By Diedre Santos Do Carmo, Marcos Piau, Israel Campiotti, Rodrigo Nogueira, Roberto de Alencar Lotufo. 2020
69 citations recorded by the source corpus.
In natural language processing (NLP), there is a need for more resources in Portuguese, since much of the data used in the state-of-the-art research is in other languages. In this paper, we pretrain a T5 model on the BrWac corpus, an extensive collection of web pages in Portuguese, and evaluate its performance against other Portuguese pretrained models and multilingual models on the sentence similarity and sentence entailment tasks. We show that our Portuguese pretrained models have significantly better performance over the original T5 models. Moreover, we showcase the positive impact of using a Portuguese vocabulary.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Colin Raffel
- Noah Constant
- Rami Al-Rfou
- Adam Roberts
- Mohit Bansal
- Mihir Kale
- Percy Liang
- Luke S. Zettlemoyer
In Computer Vision and Pattern Recognition
- Mihir Kale
- Colin Raffel
- Noah Constant
- Rami Al-Rfou
- Adam Roberts
- Ajit Rajasekharan
- Sivanesan Sangeetha
- Katikapalli Subramanyam Kalyan
In Natural Language Processing
- Colin Raffel
- Noah Constant
- Rami Al-Rfou
- Mihir Kale
- Ajit Rajasekharan
- Sivanesan Sangeetha
- Katikapalli Subramanyam Kalyan
- Aditya Siddhant
Citations in the corpus
- mT5: A Massively Multilingual Pre-trained Text-to-Text Transformer (2020, background): “…has addressed this English-centricity has been to release dozens of models that have instead been pre-trained on a single non-English language (Carmo et al., 2020; de Vries et al., 2019; Le et al., 2019; Martin et al., 2019; Delobelle et al., 2020; Malmsten et al., 2020; Nguyen and Nguyen,…”
- AMMUS : A Survey of Transformer-based Pretrained Models in Natural Language Processing (2021, methodology): “PTT5 [77] Portuguese T5 Seq2SeqLM BrWac corpus SentencePiece (32K)”