Bert Large Portuguese Cased Legal Tsdae Gpl Nli Sts MetaKD V0

🚀 stjiris/bert-large-portuguese-cased-legal-tsdae-gpl-nli-sts-MetaKD-v0 (Legal BERTimbau)

這是一個用於法律領域的葡萄牙語BERT模型，它可以將句子和段落映射到1024維的密集向量空間，適用於聚類或語義搜索等任務。

本項目是IRIS項目的一部分。

論文：A Semantic Search System for Supremo Tribunal de Justiça

✨ 主要特性

• 基於sentence-transformers，可將句子和段落映射到1024維的密集向量空間。
• 源自stjiris/bert-large-portuguese-cased-legal-tsdae（BERTimbau large的法律變體）。
• 使用TSDAE技術進行訓練，學習率為1e-5，使用約30000份文檔中的法律句子，進行了212k個訓練步驟。
• 經過生成式偽標籤（Generative Pseudo Labeling）訓練。
• 引入了NLI數據進行訓練，批次大小為16，學習率為2e-5。
• 針對語義文本相似度進行訓練，並使用assin、assin2、stsb_multi_mt pt數據集進行微調。
• 經過元數據知識蒸餾（Metadata Knowledge Distillation）處理。

模型指標

📦 安裝指南

使用此模型前，需要安裝sentence-transformers：

pip install -U sentence-transformers

💻 使用示例

基礎用法（Sentence-Transformers）

from sentence_transformers import SentenceTransformer
sentences = ["Isto é um exemplo", "Isto é um outro exemplo"]

model = SentenceTransformer('stjiris/bert-large-portuguese-cased-legal-tsdae-gpl-nli-sts-MetaKD-v0')
embeddings = model.encode(sentences)
print(embeddings)

高級用法（HuggingFace Transformers）

from transformers import AutoTokenizer, AutoModel
import torch

# Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0] # First element of model_output contains all token embeddings
    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)

# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('stjiris/bert-large-portuguese-cased-legal-tsdae-gpl-nli-sts-MetaKD-v0')
model = AutoModel.from_pretrained('stjiris/bert-large-portuguese-cased-legal-tsdae-gpl-nli-sts-MetaKD-v0')

# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')

# Compute token embeddings
with torch.no_grad():
    model_output = model(**encoded_input)
# Perform pooling. In this case, mean pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)

📚 詳細文檔

完整模型架構

SentenceTransformer(
  (0): Transformer({'max_seq_length': 514, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 1028, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False})
)

📄 許可證

本模型使用MIT許可證。

📖 引用與作者

貢獻者

@rufimelo99

如果使用本項目，請引用以下文獻：

@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"
}

@inproceedings{souza2020bertimbau,
  author    = {F{\'a}bio Souza and
               Rodrigo Nogueira and
               Roberto Lotufo},
  title     = {{BERT}imbau: pretrained {BERT} models for {B}razilian {P}ortuguese},
  booktitle = {9th Brazilian Conference on Intelligent Systems, {BRACIS}, Rio Grande do Sul, Brazil, October 20-23 (to appear)},
  year      = {2020}
}

@inproceedings{fonseca2016assin,
  title={ASSIN: Avaliacao de similaridade semantica e inferencia textual},
  author={Fonseca, E and Santos, L and Criscuolo, Marcelo and Aluisio, S},
  booktitle={Computational Processing of the Portuguese Language-12th International Conference, Tomar, Portugal},
  pages={13--15},
  year={2016}
}

@inproceedings{real2020assin,
  title={The assin 2 shared task: a quick overview},
  author={Real, Livy and Fonseca, Erick and Oliveira, Hugo Goncalo},
  booktitle={International Conference on Computational Processing of the Portuguese Language},
  pages={406--412},
  year={2020},
  organization={Springer}
}
@InProceedings{huggingface:dataset:stsb_multi_mt,
title = {Machine translated multilingual STS benchmark dataset.},
author={Philip May},
year={2021},
url={https://github.com/PhilipMay/stsb-multi-mt}
}