README.md

---
pretty_name: ExtraGLUE
language:
- pt
source_datasets:
- glue
- superglue
license: mit
viewer: true
task_categories:
- text-classification
- sentence-similarity
- question-answering
task_ids:
- language-modeling
- multi-class-classification
- natural-language-inference
- sentiment-classification
- semantic-similarity-scoring
- semantic-similarity-classification
configs:
- config_name: axb_pt-BR
  data_files:
  - split: test
    path:
    - "data/axb_pt-BR/test.jsonl"
- config_name: axb_pt-PT
  data_files:
  - split: test
    path:
    - "data/axb_pt-PT/test.jsonl"
- config_name: axg_pt-BR
  data_files:
  - split: test
    path:
    - "data/axg_pt-BR/test.jsonl"
- config_name: axg_pt-PT
  data_files:
  - split: test
    path:
    - "data/axg_pt-PT/test.jsonl"
- config_name: boolq_pt-BR
  data_files:
  - split: train
    path:
    - "data/boolq_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/boolq_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/boolq_pt-BR/test.jsonl"
- config_name: boolq_pt-PT
  data_files:
  - split: train
    path:
    - "data/boolq_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/boolq_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/boolq_pt-PT/test.jsonl"
- config_name: cb_pt-BR
  data_files:
  - split: train
    path:
    - "data/cb_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/cb_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/cb_pt-BR/test.jsonl"
- config_name: cb_pt-PT
  data_files:
  - split: train
    path:
    - "data/cb_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/cb_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/cb_pt-PT/test.jsonl"
- config_name: copa_pt-BR
  data_files:
  - split: train
    path:
    - "data/copa_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/copa_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/copa_pt-BR/test.jsonl"
- config_name: copa_pt-PT
  data_files:
  - split: train
    path:
    - "data/copa_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/copa_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/copa_pt-PT/test.jsonl"
- config_name: mnli_matched_pt-BR
  data_files:
  - split: train
    path:
    - "data/mnli_matched_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/mnli_matched_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/mnli_matched_pt-BR/test.jsonl"
- config_name: mnli_matched_pt-PT
  data_files:
  - split: train
    path:
    - "data/mnli_matched_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/mnli_matched_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/mnli_matched_pt-PT/test.jsonl"
- config_name: mnli_mismatched_pt-BR
  data_files:
  - split: train
    path:
    - "data/mnli_mismatched_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/mnli_mismatched_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/mnli_mismatched_pt-BR/test.jsonl"
- config_name: mnli_mismatched_pt-PT
  data_files:
  - split: train
    path:
    - "data/mnli_mismatched_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/mnli_mismatched_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/mnli_mismatched_pt-PT/test.jsonl"
- config_name: mrpc_pt-BR
  data_files:
  - split: train
    path:
    - "data/mrpc_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/mrpc_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/mrpc_pt-BR/test.jsonl"
- config_name: mrpc_pt-PT
  data_files:
  - split: train
    path:
    - "data/mrpc_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/mrpc_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/mrpc_pt-PT/test.jsonl"
- config_name: multirc_pt-BR
  data_files:
  - split: train
    path:
    - "data/multirc_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/multirc_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/multirc_pt-BR/test.jsonl"
- config_name: multirc_pt-PT
  data_files:
  - split: train
    path:
    - "data/multirc_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/multirc_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/multirc_pt-PT/test.jsonl"
- config_name: qnli_pt-BR
  data_files:
  - split: train
    path:
    - "data/qnli_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/qnli_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/qnli_pt-BR/test.jsonl"
- config_name: qnli_pt-PT
  data_files:
  - split: train
    path:
    - "data/qnli_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/qnli_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/qnli_pt-PT/test.jsonl"
- config_name: rte_pt-BR
  data_files:
  - split: train
    path:
    - "data/rte_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/rte_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/rte_pt-BR/test.jsonl"
- config_name: rte_pt-PT
  data_files:
  - split: train
    path:
    - "data/rte_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/rte_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/rte_pt-PT/test.jsonl"
- config_name: sst2_pt-BR
  data_files:
  - split: train
    path:
    - "data/sst2_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/sst2_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/sst2_pt-BR/test.jsonl"
- config_name: sst2_pt-PT
  data_files:
  - split: train
    path:
    - "data/sst2_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/sst2_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/sst2_pt-PT/test.jsonl"
- config_name: stsb_pt-BR
  data_files:
  - split: train
    path:
    - "data/stsb_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/stsb_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/stsb_pt-BR/test.jsonl"
- config_name: stsb_pt-PT
  data_files:
  - split: train
    path:
    - "data/stsb_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/stsb_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/stsb_pt-PT/test.jsonl"
- config_name: wnli_pt-BR
  data_files:
  - split: train
    path:
    - "data/wnli_pt-BR/train.jsonl"
  - split: validation
    path:
    - "data/wnli_pt-BR/validation.jsonl"
  - split: test
    path:
    - "data/wnli_pt-BR/test.jsonl"
- config_name: wnli_pt-PT
  data_files:
  - split: train
    path:
    - "data/wnli_pt-PT/train.jsonl"
  - split: validation
    path:
    - "data/wnli_pt-PT/validation.jsonl"
  - split: test
    path:
    - "data/wnli_pt-PT/test.jsonl"
---

</br>
</br>
<img align="left" width="40" height="40" src="https://github.githubassets.com/images/icons/emoji/unicode/1f917.png">
<p style="text-align: center;">&nbsp;&nbsp;&nbsp;&nbsp;This is the dataset card for extraGLUE. 
  You may be interested in some of the other <a href="https://huggingface.co/PORTULAN">datasets for Portuguese</a> and in the models trained with them, 
  namely <a href="https://huggingface.co/PORTULAN">Albertina (encoders) and Gervásio (decoders) families</a>.
</p>
</br>
</br>


ExtraGLUE
===
</br>


ExtraGLUE is a Portuguese dataset obtained by the automatic translation of some of the tasks in the GLUE and SuperGLUE benchmarks.
Two variants of Portuguese are considered, namely European Portuguese and American Portuguese.
The dataset is distributed for free under an open license.

The 14 tasks in extraGLUE cover different aspects of language understanding: 

*Single sentence*

- **SST-2** is a task for predicting the sentiment polarity of movie reviews.

*Semantic similarity*

- **MRPC** is a task for determining whether a pair of sentences are mutual paraphrases.
- **STS-B** is a task for predicting a similarity score (from 1 to 5) for each sentence pair.

*Inference*

- **MNLI** is a task to determine if a given premise sentence entails, contradicts, or is neutral to a hypothesis sentence; this task includes **matched** (in-domain) and **mismatched** (cross-domain) validation and test sets.
- **QNLI** is a question-answering task converted to determine whether the context sentence contains the answer to the question.
- **RTE** is a task for determining whether a premise sentence entails a hypothesis sentence.
- **WNLI** is a pronoun resolution task formulated as sentence pair entailment classification where, in the second sentence, the pronoun is replaced by a possible referent.
- **CB** comprises short texts with embedded clauses; one such clause is extracted as a hypothesis and should be classified as neutral, entailment or contradiction.
- **AX_b** is designed to test models across a wide spectrum of linguistic, commonsense, and world knowledge; each instance contains a sentence pair labeled with entailment or not entailment.
- **AX_g** is designed to measure gender bias, where each premise sentence includes a male or female pronoun and a hypothesis includes a possible referent for the pronoun.

*Question answering*

- **BoolQ** is a question-answering task where yes/no questions are given for short text passages.
- **MultiRC** is a task where, given a context paragraph, a question, and an answer, the goal is to determine whether the answer is true; for the same context and question, more than one answer may be correct.

*Reasoning*

- **CoPA** is a casual reasoning task: given a premise, two choices, and a cause/effect prompt, the system must choose one of the choices.

If you use this dataset please cite:

    @inproceedings{osorio-etal-2024-portulan,
      title = "{PORTULAN} {E}xtra{GLUE} Datasets and Models: Kick-starting a Benchmark for the Neural Processing of {P}ortuguese",
      author = "Os{\'o}rio, Tom{\'a}s Freitas  and
        Leite, Bernardo  and
        Lopes Cardoso, Henrique  and
        Gomes, Lu{\'\i}s  and
        Rodrigues, Jo{\~a}o  and
        Santos, Rodrigo  and
        Branco, Ant{\'o}nio",
      editor = "Zweigenbaum, Pierre  and
        Rapp, Reinhard  and
        Sharoff, Serge",
      booktitle = "Proceedings of the 17th Workshop on Building and Using Comparable Corpora (BUCC) @ LREC-COLING 2024",
      month = may,
      year = "2024",
      address = "Torino, Italia",
      publisher = "ELRA and ICCL",
      url = "https://aclanthology.org/2024.bucc-1.3",
      pages = "24--34",
    }

# Acknowledgments

The research reported here was partially supported by:
PORTULAN CLARIN—Research Infrastructure for the Science and Technology of Language, funded by Lisboa 2020, Alentejo 2020 and FCT—Fundação para a Ciência e Tecnologia under the
grant PINFRA/22117/2016;
research project GPT-PT - Transformer-based Decoder for the Portuguese Language, funded by FCT—Fundação para a Ciência e Tecnologia under the
grant CPCA-IAC/AV/478395/2022;
innovation project ACCELERAT.AI - Multilingual Intelligent Contact Centers, funded by IAPMEI, I.P. - Agência para a Competitividade e Inovação 
under the grant C625734525-00462629, of Plano de Recuperação e Resiliência, call RE-C05-i01.01 – Agendas/Alianças Mobilizadoras para a Reindustrialização;
and Base Funding (UIDB/00027/2020) and Programmatic Funding (UIDP/00027/2020) of the Artificial Intelligence and Computer Science Laboratory (LIACC) funded by national funds through FCT/MCTES (PIDDAC).