INESC-ID

A Semantic Search System for Supremo Tribunal de Justiça

Work developed as part of Project IRIS.

Thesis: A Semantic Search System for Supremo Tribunal de Justiça

stjiris/t5-portuguese-legal-summarization

T5 Model fine-tuned over “unicamp-dl/ptt5-base-portuguese-vocab” t5 model.

We utilized various jurisprudence and its summary to train this model.

Usage (HuggingFace transformers)

# name of folder principal
from transformers import T5Tokenizer, T5ForConditionalGeneration

model_checkpoint = "stjiris/t5-portuguese-legal-summarization"
t5_model = T5ForConditionalGeneration.from_pretrained(model_checkpoint)
t5_tokenizer = T5Tokenizer.from_pretrained(model_checkpoint)

preprocess_text = "These are some big words and text and words and text, again, that we want to summarize"
t5_prepared_Text = "summarize: "+preprocess_text
#print ("original text preprocessed: \n", preprocess_text)

tokenized_text = t5_tokenizer.encode(t5_prepared_Text, return_tensors="pt").to(device)


# summmarize 
summary_ids = t5_model.generate(tokenized_text,
                                    num_beams=4,
                                    no_repeat_ngram_size=2,
                                    min_length=512,
                                    max_length=1024,
                                    early_stopping=True)

output = t5_tokenizer.decode(summary_ids[0], skip_special_tokens=True)

print ("\n\nSummarized text: \n",output)

Citing & Authors

Contributions

@rufimelo99

If you use this work, please cite:

@InProceedings{MeloSemantic,
  author="Melo, Rui
  and Santos, Pedro A.
  and Dias, Jo{\~a}o",
  editor="Moniz, Nuno
  and Vale, Zita
  and Cascalho, Jos{\'e}
  and Silva, Catarina
  and Sebasti{\~a}o, Raquel",
  title="A Semantic Search System for the Supremo Tribunal de Justi{\c{c}}a",
  booktitle="Progress in Artificial Intelligence",
  year="2023",
  publisher="Springer Nature Switzerland",
  address="Cham",
  pages="142--154",
  abstract="Many information retrieval systems use lexical approaches to retrieve information. Such approaches have multiple limitations, and these constraints are exacerbated when tied to specific domains, such as the legal one. Large language models, such as BERT, deeply understand a language and may overcome the limitations of older methodologies, such as BM25. This work investigated and developed a prototype of a Semantic Search System to assist the Supremo Tribunal de Justi{\c{c}}a (Portuguese Supreme Court of Justice) in its decision-making process. We built a Semantic Search System that uses specially trained BERT models (Legal-BERTimbau variants) and a Hybrid Search System that incorporates both lexical and semantic techniques by combining the capabilities of BM25 and the potential of Legal-BERTimbau. In this context, we obtained a {\$}{\$}335{\backslash}{\%}{\$}{\$}335{\%}increase on the discovery metric when compared to BM25 for the first query result. This work also provides information on the most relevant techniques for training a Large Language Model adapted to Portuguese jurisprudence and introduces a new technique of Metadata Knowledge Distillation.",
  isbn="978-3-031-49011-8"
}


@article{ptt5_2020,
  title={PTT5: Pretraining and validating the T5 model on Brazilian Portuguese data},
  author={Carmo, Diedre and Piau, Marcos and Campiotti, Israel and Nogueira, Rodrigo and Lotufo, Roberto},
  journal={arXiv preprint arXiv:2008.09144},
  year={2020}
}
Downloads last month
333
Safetensors
Model size
223M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Space using stjiris/t5-portuguese-legal-summarization 1