CoChat
Sign Up
Cited by 317Natural Language Processing TechniquesTopic ModelingComputational and Text Analysis Methods

The DOI matches its Crossref record; 67 of the 70 references it cites were found in a registry or an index.

Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus

Open access
Dodge, Jesse, Sap, Maarten, Marasović, Ana, Agnew, William, Ilharco, Gabriel, Groeneveld, Dirk, Mitchell, Margaret, Gardner, Matt·2021·Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
DOI 10.18653/v1/2021.emnlp-main.98
Open this paper in CoChat — free accountOpen PDF

Abstract

Jesse Dodge, Maarten Sap, Ana Marasović, William Agnew, Gabriel Ilharco, Dirk Groeneveld, Margaret Mitchell, Matt Gardner. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 2021.

Overview

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

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

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

Integrity signals

Registry

Crossref · 10.18653/v1/2021.emnlp-main.98

Cites

67 of 70 references verified0 citation errors · 3 not checkable

Statistics

Figures

Figure 1 We advocate for three levels of documentation when creating web-crawled corpora. On the right, we include some example of types of documentation that we provide for the C4.EN dataset.
Figure 1 We advocate for three levels of documentation when creating web-crawled corpora. On the right, we include some example of types of documentation that we provide for the C4.EN dataset.CC-BY · From the PDF
Figure 2 Number of tokens from the 25 most represented top-level domains (left) and websites (right) in C4.EN.
Figure 2 Number of tokens from the 25 most represented top-level domains (left) and websites (right) in C4.EN.CC-BY · From the PDF
Figure 3 The date URLs were first indexed by the Internet Archive 13 before the Common Crawl snapshot was collected.
Figure 3 The date URLs were first indexed by the Internet Archive 13 before the Common Crawl snapshot was collected.CC-BY · From the PDF
Figure 4 URL frequency by country for 175,000 randomly selected URLS from the cleaned common crawl dataset.
Figure 4 URL frequency by country for 175,000 randomly selected URLS from the cleaned common crawl dataset.CC-BY · From the PDF
Figure 5 Pointwise Mutual Information (PMI) between identity mentions and documents being filtered out by the blocklist. Identities with higher PMI (e.g., lesbian, gay) have higher likelihood of being filtered out.
Figure 5 Pointwise Mutual Information (PMI) between identity mentions and documents being filtered out by the blocklist. Identities with higher PMI (e.g., lesbian, gay) have higher likelihood of being filtered out.CC-BY · From the PDF
Figure 6 K-means clustering of 100k randomly sampled filtered documents encoded using TF-IDF and tSNE PCA (only 5k shown for clarity). Five top keywords for each cluster given in legend.
Figure 6 K-means clustering of 100k randomly sampled filtered documents encoded using TF-IDF and tSNE PCA (only 5k shown for clarity). Five top keywords for each cluster given in legend.CC-BY · From the PDF
Table 6: List of regular expressions used to capture the identity mentions studied in §5.2
Table 6: List of regular expressions used to capture the identity mentions studied in §5.2CC-BY · From the PDF

Verification

Claims the paper makes

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

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

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 65 references

Research works printed but not linked: 7

  1. Isaac Caswell, Julia Kreutzer, Lisa Wang, Ahsan Wa- hab, D. V. Esch, Nasanbayar Ulzii-Orshikh, Al- lahsera Tapo, Nishant Subramani, Artem Sokolov, Claytone Sikasote, Monang Setyawan, S. Sarin, Sokhar Samb, B. Sagot, C. Rivera, Annette Rios Gonzales, Isabel Papadimitriou, S. Osei, Pedro Javier Ortiz Suárez, Iroro Orife, Kelechi Ogueji, Rubungo Andre Niyongabo, Toan Q. Nguyen, Math- ias Muller, A. Muller, S. Muhammad, N. Muham- mad, Ayanda Mnyakeni, Jamshidbek Mirzakhalov, Tapiwanashe Matangira, Colin Leong, Nze Lawson, Sneha Kudugunta, Yacine Jernite, M. Jenny, Orhan Firat, Bonaventure F. P. Dossou, Sakhile Dlamini, N. D. Silva, Sakine cCabuk Balli, Stella Rose Biderman, Alessia Battisti, A. Baruwa, Ankur Bapna, Pallavi Baljekar, Israel Abebe Azime, Ayo- dele Awokoya, Duygu Ataman, Orevaoghene Ahia, Oghenefego Ahia, Sweta Agrawal, and Mofetoluwa Adeyemi. 2021. Quality at a glance: An audit of web-crawled multilingual datasets. In Proceedings of the AfricanNLP Workshop.
  2. Kenneth Church and Patrick Hanks. 1990. Word as- sociation norms, mutual information, and lexicogra- phy. Computational linguistics, 16(1):22-29.
  3. Danilo Giampiccolo, Bernardo Magnini, Ido Dagan, and Bill Dolan. 2007. The third PASCAL recogniz- ing textual entailment challenge. In Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing, pages 1-9, Prague. Association for Computational Linguistics.
  4. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text trans- former. Journal of Machine Learning Research.
  5. Jonathan Rosa. 2019. Looking like a language, sound- ing like a race. Oxford University Press. · DOI
  6. Rachael Tatman. 2020. What i won't build. WiNLP Workshop at ACL.
  7. Rowan Zellers, Ari Holtzman, Hannah Rashkin, Yonatan Bisk, Ali Farhadi, Franziska Roesner, and Yejin Choi. 2019. Defending against neural fake news. In NeurIPS.

Cited by

Sorted by citations

Items 201–300 of 317

Related

See all

Metadata

Authors
Dodge, J., Sap, M., Marasović, A., Agnew, W., Ilharco, G., Groeneveld, D., Mitchell, M., & Gardner, M.
Affiliations
University of Washington (US)
ORCID
Venue
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Publisher
Association for Computational Linguistics
Published
2021
Pages
1286-1305
DOI
10.18653/v1/2021.emnlp-main.98
OpenAlex
W3204096139
Cited by
306
References
65
Citations by year
2026: 76 · 2025: 77 · 2024: 54 · 2023: 65 · 2022: 33 · 2021: 1
FWCI
17.67
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) (cc-by)
Funders
Allen Institute for Artificial Intelligence
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
  • Social Sciences › General Social Sciences › Computational and Text Analysis Methods
Keywords
web crawling, corpus annotation

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