Edit model card


RoBERTaLexPT-base is a Portuguese Masked Language Model pretrained from scratch from the LegalPT and CrawlPT corpora, using the same architecture as RoBERTa-base, introduced by Liu et al. (2019).


The model was evaluated on "PortuLex" benchmark, a four-task benchmark designed to evaluate the quality and performance of language models in the Portuguese legal domain.

Macro F1-Score (%) for multiple models evaluated on PortuLex benchmark test splits:

Model LeNER UlyNER-PL FGV-STF RRIP Average (%)
Coarse/Fine Coarse
BERTimbau-base 88.34 86.39/83.83 79.34 82.34 83.78
BERTimbau-large 88.64 87.77/84.74 79.71 83.79 84.60
Albertina-PT-BR-base 89.26 86.35/84.63 79.30 81.16 83.80
Albertina-PT-BR-xlarge 90.09 88.36/86.62 79.94 82.79 85.08
BERTikal-base 83.68 79.21/75.70 77.73 81.11 79.99
JurisBERT-base 81.74 81.67/77.97 76.04 80.85 79.61
BERTimbauLAW-base 84.90 87.11/84.42 79.78 82.35 83.20
Legal-XLM-R-base 87.48 83.49/83.16 79.79 82.35 83.24
Legal-XLM-R-large 88.39 84.65/84.55 79.36 81.66 83.50
Legal-RoBERTa-PT-large 87.96 88.32/84.83 79.57 81.98 84.02
RoBERTaTimbau-base (Reproduction of BERTimbau) 89.68 87.53/85.74 78.82 82.03 84.29
RoBERTaLegalPT-base (Trained on LegalPT) 90.59 85.45/84.40 79.92 82.84 84.57
RoBERTaCrawlPT-base (Trained on CrawlPT) 89.24 88.22/86.58 79.88 82.80 84.83
RoBERTaLexPT-base (this) (Trained on CrawlPT + LegalPT) 90.73 88.56/86.03 80.40 83.22 85.41

In summary, RoBERTaLexPT consistently achieves top legal NLP effectiveness despite its base size. With sufficient pre-training data, it can surpass larger models. The results highlight the importance of domain-diverse training data over sheer model scale.

Training Details

RoBERTaLexPT-base is pretrained on:

Training Procedure

Our pretraining process was executed using the Fairseq library v0.10.2 on a DGX-A100 cluster, utilizing a total of 2 Nvidia A100 80 GB GPUs. The complete training of a single configuration takes approximately three days.

This computational cost is similar to the work of BERTimbau-base, exposing the model to approximately 65 billion tokens during training.


We deduplicated all subsets of the LegalPT and CrawlPT Corpus using the a MinHash algorithm and Locality Sensitive Hashing implementation from the libary text-dedup to find clusters of duplicate documents.

To ensure that domain models are not constrained by a generic vocabulary, we utilized the HuggingFace Tokenizers -- BPE algorithm to train a vocabulary for each pre-training corpus used.

Training Hyperparameters

The pretraining process involved training the model for 62,500 steps, with a batch size of 2048 and a learning rate of 4e-4, each sequence containing a maximum of 512 tokens.
The weight initialization is random.
We employed the masked language modeling objective, where 15% of the input tokens were randomly masked.
The optimization was performed using the AdamW optimizer with a linear warmup and a linear decay learning rate schedule.

For other parameters we adopted the standard RoBERTa-base hyperparameters:

Hyperparameter RoBERTa-base
Number of layers 12
Hidden size 768
FFN inner hidden size 3072
Attention heads 12
Attention head size 64
Dropout 0.1
Attention dropout 0.1
Warmup steps 6k
Peak learning rate 4e-4
Batch size 2048
Weight decay 0.01
Maximum training steps 62.5k
Learning rate decay Linear
AdamW $$\epsilon$$ 1e-6
AdamW $$\beta_1$$ 0.9
AdamW $$\beta_2$$ 0.98
Gradient clipping 0.0


    title = "{R}o{BERT}a{L}ex{PT}: A Legal {R}o{BERT}a Model pretrained with deduplication for {P}ortuguese",
    author = "Garcia, Eduardo A. S.  and
      Silva, Nadia F. F.  and
      Siqueira, Felipe  and
      Albuquerque, Hidelberg O.  and
      Gomes, Juliana R. S.  and
      Souza, Ellen  and
      Lima, Eliomar A.",
    editor = "Gamallo, Pablo  and
      Claro, Daniela  and
      Teixeira, Ant{\'o}nio  and
      Real, Livy  and
      Garcia, Marcos  and
      Oliveira, Hugo Gon{\c{c}}alo  and
      Amaro, Raquel",
    booktitle = "Proceedings of the 16th International Conference on Computational Processing of Portuguese",
    month = mar,
    year = "2024",
    address = "Santiago de Compostela, Galicia/Spain",
    publisher = "Association for Computational Lingustics",
    url = "https://aclanthology.org/2024.propor-1.38",
    pages = "374--383",


This work has been supported by the AI Center of Excellence (Centro de Excelência em Inteligência Artificial – CEIA) of the Institute of Informatics at the Federal University of Goiás (INF-UFG).

Downloads last month
Model size
125M params
Tensor type

Datasets used to train eduagarcia/RoBERTaLexPT-base

Space using eduagarcia/RoBERTaLexPT-base 1

Collection including eduagarcia/RoBERTaLexPT-base

Evaluation results