Hugging Face's logo

eduagarcia
/
RoBERTaLexPT-base

Fill-Mask
Transformers
PyTorch
Safetensors
Portuguese
roberta
legal
Eval Results
Inference Endpoints
Model card Files Community
6
RoBERTaLexPT-base / README.md
eduagarcia's picture
eduagarcia
Update
a197b92 verified
preview code
|
raw
history blame
No virus
8.36 kB
metadata
datasets:
  - eduagarcia/LegalPT
  - eduagarcia/cc100-pt
  - eduagarcia/OSCAR-2301-pt_dedup
  - eduagarcia/brwac_dedup
language:
  - pt
pipeline_tag: fill-mask
tags:
  - legal
model-index:
  - name: RoBERTaLexPT-base
    results:
      - task:
          type: token-classification
        dataset:
          type: eduagarcia/portuguese_benchmark
          name: LeNER
          config: LeNER-Br
          split: test
        metrics:
          - type: seqeval
            value: 90.73
            name: Mean F1
            args:
              scheme: IOB2
      - task:
          type: token-classification
        dataset:
          type: eduagarcia/portuguese_benchmark
          name: UlyNER-PL Coarse
          config: UlyssesNER-Br-PL-coarse
          split: test
        metrics:
          - type: seqeval
            value: 88.56
            name: Mean F1
            args:
              scheme: IOB2
      - task:
          type: token-classification
        dataset:
          type: eduagarcia/portuguese_benchmark
          name: UlyNER-PL Fine
          config: UlyssesNER-Br-PL-fine
          split: test
        metrics:
          - type: seqeval
            value: 86.03
            name: Mean F1
            args:
              scheme: IOB2
license: cc-by-4.0
metrics:
  - seqeval

RoBERTaLexPT-base

RoBERTaLexPT-base is pretrained from LegalPT and CrawlPT corpora, using RoBERTa-base, introduced by Liu et al. (2019).

Evaluation

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 89.68 87.53/85.74 78.82 82.03 84.29
RoBERTaLegalPT-base 90.59 85.45/84.40 79.92 82.84 84.57
RoBERTaLexPT-base 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 overparameterized models. The results highlight the importance of domain-diverse training data over sheer model scale.

Training Details

RoBERTaLexPT-base is pretrained from both data:

  • LegalPT is a Portuguese legal corpus by aggregating diverse sources of up to 125GiB data.
  • CrawlPT is a duplication of three Portuguese general corpora: brWaC, CC100-PT, OSCAR-2301.

Training Procedure

Our pretraining process was executed using the Fairseq library 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 setup is similar to the work of BERTimbau, exposing the model to approximately 65 billion tokens during training.

Preprocessing

Following the approach of Lee et al. (2022), we deduplicated all subsets of the LegalPT Corpus using the MinHash algorithm and Locality Sensitive Hashing 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 sequences, each containing a maximum of 512 tokens. 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 hyperparameters we adopted the standard RoBERTa 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

Citation 

@InProceedings{garcia2024_roberlexpt,
    author="Garcia, Eduardo A. S.
    and Silva, N{\'a}dia F. F.
    and Siqueira, Felipe
    and Gomes, Juliana R. S.
    and Albuqueruqe, Hidelberg O.
    and Souza, Ellen
    and Lima, Eliomar
    and De Carvalho, André",
    title="RoBERTaLexPT: A Legal RoBERTa Model pretrained with deduplication for Portuguese",
    booktitle="Computational Processing of the Portuguese Language",
    year="2024",
    publisher="Association for Computational Linguistics"
}

Acknowledgment

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).