natural_questions_open

  • Descrição :

A tarefa NQ-Open, introduzida por Lee et.al. 2019, é um benchmark de resposta a perguntas de domínio aberto derivado de Natural Questions. O objetivo é prever uma string de resposta em inglês para uma pergunta de entrada em inglês. Todas as perguntas podem ser respondidas usando o conteúdo da Wikipédia em inglês.

Dividir Exemplos
'train' 87.925
'validation' 3.610
  • Estrutura de recursos :
FeaturesDict({
    'answer': Sequence(string),
    'question': string,
})
  • Documentação do recurso:
Recurso Aula Forma Tipo D Descrição
RecursosDict
responder Sequência(Tensor) (Nenhum,) corda
pergunta tensor corda
  • Citação :
@inproceedings{orqa,
title = {Latent Retrieval for Weakly Supervised Open Domain Question Answering},
author = {Lee, Kenton and Chang, Ming-Wei and Toutanova, Kristina},
year = {2019},
month = {01},
pages = {6086-6096},
booktitle = {Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics},
doi = {10.18653/v1/P19-1612}
}

@article{47761,
title = {Natural Questions: a Benchmark for Question Answering Research},
author = {Tom Kwiatkowski and Jennimaria Palomaki and Olivia Redfield and Michael Collins and Ankur Parikh and Chris Alberti and Danielle Epstein and Illia Polosukhin and Matthew Kelcey and Jacob Devlin and Kenton Lee and Kristina N. Toutanova and Llion Jones and Ming-Wei Chang and Andrew Dai and Jakob Uszkoreit and Quoc Le and Slav Petrov},
year = {2019},
journal = {Transactions of the Association of Computational Linguistics}
}