@InProceedings{wenzek-EtAl:2020:LREC,
  author    = {Wenzek, Guillaume  and  Lachaux, Marie-Anne  and  Conneau, Alexis  and  Chaudhary, Vishrav  and  Guzmán, Francisco  and  Joulin, Armand  and  Grave, Edouard},
  title     = {CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data},
  booktitle      = {Proceedings of The 12th Language Resources and Evaluation Conference},
  month          = {May},
  year           = {2020},
  address        = {Marseille, France},
  publisher      = {European Language Resources Association},
  pages     = {4003--4012},
  abstract  = {Pre-training text representations have led to significant improvements in many areas of natural language processing. The quality of these models benefits greatly from the size of the pretraining corpora as long as its quality is preserved. In this paper, we describe an automatic pipeline to extract massive high-quality monolingual datasets from Common Crawl for a variety of languages. Our pipeline follows the data processing introduced in fastText (Mikolov et al., 2017; Grave et al., 2018), that deduplicates documents and identifies their language. We augment this pipeline with a filtering step to select documents that are close to high quality corpora like Wikipedia.},
  url       = {https://www.aclweb.org/anthology/2020.lrec-1.494}
}