Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index.

Cloze-driven Pretraining of Self-attention Networks

Open access
Baevski, Alexei, Edunov, Sergey, Liu, Yinhan, Zettlemoyer, Luke, Auli, Michael · 2019 · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)

Abstract

Alexei Baevski, Sergey Edunov, Yinhan Liu, Luke Zettlemoyer, Michael Auli. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.18653/v1/d19-1539

Cites

37 of 37 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Illustration of the model. Block i is a standard transformer decoder block. Green blocks operate left to right by masking future time-steps and blue blocks operate right to left. At the top, states are combined with a standard multi-head self-attention module whose output is fed to a classifier that predicts the center token.
Figure 1 Illustration of the model. Block i is a standard transformer decoder block. Green blocks operate left to right by masking future time-steps and blue blocks operate right to left. At the top, states are combined with a standard multi-head self-attention module whose output is fed to a classifier that predicts the center token.CC-BY · From the PDF
Figure 2 Illustration of fine-tuning for a downstream task. For classification problems, output of the first and last token is fed to a task-specific classifier. Masking for the final combination layer (comb) is removed which results in representations based on all forward and backward states (cf. Figure 1). The red dot-dashed arrows show connections that are masked during training, but unmasked for fine-tuning.
Figure 2 Illustration of fine-tuning for a downstream task. For classification problems, output of the first and last token is fed to a task-specific classifier. Masking for the final combination layer (comb) is removed which results in representations based on all forward and backward states (cf. Figure 1). The red dot-dashed arrows show connections that are masked during training, but unmasked for fine-tuning.CC-BY · From the PDF
Figure 3 Average GLUE score with different amounts of Common Crawl data for pretraining.
Figure 3 Average GLUE score with different amounts of Common Crawl data for pretraining.CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 35 references

Research works printed but not linked: 6

  1. Danilo Giampiccolo, Bernardo Magnini, Ido Dagan, and Bill Dolan. 2007. The pascal recognizing textual entailment challenge. Proc. of the ACL- PASCAL workshop on textual entailment and para- phrasing.
  2. Edouard Grave, Piotr Bojanowski, Prakhar Gupta, Ar- mand Joulin, and Tomas Mikolov. 2018. Learning word vectors for 157 languages. In Proc. of LREC.
  3. Bryan McCann, James Bradbury, Caiming Xiong, and Richard Socher. 2017. Learned in translation: Con- textualized word vectors. In Proc. of NIPS.
  4. Myle Ott, Sergey Edunov, David Grangier, and Michael Auli. 2018. Scaling neural machine trans- lation. In Proc. of WMT.
  5. Matthew E Peters, Mark Neumann, Mohit Iyyer, Matt Gardner, Christopher Clark, Kenton Lee, and Luke Zettlemoyer. 2018. Deep contextualized word rep- resentations. In Proc. of ACL.
  6. Rico Sennrich, Barry Haddow, and Alexandra Birch. 2016. Neural machine translation of rare words with subword units. In Proc. of ACL.

Cited by

See all 234

Metadata

Authors
Baevski, A., Edunov, S., Liu, Y., Zettlemoyer, L., & Auli, M.
Affiliations
Meta (Israel) (IL)
Venue
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
Published
2019
Pages
5359-5368
OpenAlex
W2920812691
Cited by
215
References
35
Citations by year
2026: 2 · 2025: 8 · 2024: 15 · 2023: 18 · 2022: 36 · 2021: 50 · 2020: 67 · 2019: 19
FWCI
16.45
Type
conference paper
Language
en
Open access
Licence
cc-by
Locations
  • Conference on Empirical Methods in Natural Language Processing (EMNLP) (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
  • Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
  • Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
Keywords
cloze task, self-attention network, pre-training, language model pretraining, language modeling

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Cloze-driven Pretraining of Self-attention Networks. https://app.cochat.ai/papers/10.18653/v1/d19-1539
The paper
Baevski, A., Edunov, S., Liu, Y., Zettlemoyer, L., & Auli, M. (2019). Cloze-driven Pretraining of Self-attention Networks. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). https://doi.org/10.18653/v1/d19-1539

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.