multi_nli

  • Descrizione :

Il corpus Multi-Genre Natural Language Inference (MultiNLI) è una raccolta crowd-sourced di 433.000 coppie di frasi annotate con informazioni di implicazione testuale. Il corpus è modellato sul corpus SNLI, ma differisce in quanto copre una gamma di generi di testo parlato e scritto e supporta una valutazione distintiva di generalizzazione tra generi. Il corpus è servito come base per il compito condiviso del workshop RepEval 2017 all'EMNLP di Copenaghen.

Diviso Esempi
'train' 392.702
'validation_matched' 9.815
'validation_mismatched' 9.832
  • Struttura delle caratteristiche :
FeaturesDict({
   
'hypothesis': Text(shape=(), dtype=string),
   
'label': ClassLabel(shape=(), dtype=int64, num_classes=3),
   
'premise': Text(shape=(), dtype=string),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
ipotesi Testo corda
etichetta ClassLabel int64
premessa Testo corda
  • Citazione :
@InProceedings{N18-1101,
  author
= "Williams, Adina
            and Nangia, Nikita
            and Bowman, Samuel"
,
  title
= "A Broad-Coverage Challenge Corpus for
           Sentence Understanding through Inference"
,
  booktitle
= "Proceedings of the 2018 Conference of
               the North American Chapter of the
               Association for Computational Linguistics:
               Human Language Technologies, Volume 1 (Long
               Papers)"
,
  year
= "2018",
  publisher
= "Association for Computational Linguistics",
  pages
= "1112--1122",
  location
= "New Orleans, Louisiana",
  url
= "http://aclweb.org/anthology/N18-1101"
}