Data Statements for Natural Language Processing: Toward Mitigating System Bias and Enabling Better Science
Open accessThe DOI matches its Crossref record; 46 of the 46 references it cites were found in a registry or an index.
Abstract
In this paper, we propose data statements as a design solution and professional practice for natural language processing technologists, in both research and development. Through the adoption and widespread use of data statements, the field can begin to address critical scientific and ethical issues that result from the use of data from certain populations in the development of technology for other populations. We present a form that data statements can take and explore the implications of adopting them as part of regular practice. We argue that data statements will help alleviate issues related to exclusion and bias in language technology, lead to better precision in claims about how natural language processing research can generalize and thus better engineering results, protect companies from public embarrassment, and ultimately lead to language technology that meets its users in their own preferred linguistic style and furthermore does not misrepresent them to others.
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.
Integrity signals
Registry
Cites
Statistics
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.
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?
The cited source says what the claim attributes to it.
In the citing paper
“or lead to adverse effects from deploying suboptimal or harmful models (Bender and Friedman, 2018) on tasks for which they are ill-suited (Raji et al., 2022).”
From Lessons from the Trenches on Reproducible Evaluation of Language Models (2024) · cites this paper as Bender and Friedman, 2018
The cited source says what the claim attributes to it.
In the citing paper
“Numerous authors have observed that ML systems perform better for users whose demographic characteristics match those represented in the training data [9,36].”
From A Survey of Data Quality Requirements That Matter in ML Development Pipelines (2023) · cites this paper as [9]
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
Moher, D., Hopewell, S., Schulz, K. F., Montori, V., Gotzsche, P. C., Devereaux, P. J., Elbourne, D., Egger, M., & Altman, D. G. (2010) · BMJ · doi:10.1136/bmj.c869 · Cited by 8,883
Building a Large Annotated Corpus of English: The Penn Treebank
Marcus, M. (1993) · doi:10.21236/ada273556 · Cited by 7,405
Thumbs up?: sentiment classification using machine learning techniques
Pang, B., Lee, L., & Vaithyanathan, S. (2002) · Proceedings of the ACL-02 conference on Empirical methods in natural language processing - EMNLP '02 · doi:10.3115/1118693.1118704 · Cited by 7,024 · cited 1×
“Here, our focus is on linguistic data. the review (e.g., Pang et al., 2002) or the hashtag #sarcasm used to identify sarcastic language (e.g., Kreuz and Caucci, 2007).”
BackgroundThe Study of Second Language Acquisition
Brown, C., & Ellis, R. (1996) · The Modern Language Journal · doi:10.2307/329065 · Cited by 4,199
Publication Manual of the American Psychological Association
DeCleene, K. E., & Fogo, J. (2012) · Occupational Therapy In Health Care · doi:10.3109/07380577.2011.629024 · Cited by 3,167
The Social Stratification of English in New York City.
Crockett, H. J., & Labov, W. (1968) · American Sociological Review · doi:10.2307/2092899 · Cited by 3,084
Accurate unlexicalized parsing
Klein, D., & Manning, C. D. (2003) · Proceedings of the 41st Annual Meeting on Association for Computational Linguistics - ACL '03 · doi:10.3115/1075096.1075150 · Cited by 3,008
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., Daumé, H., & Crawford, K. (2018) · arXiv · doi:10.48550/arxiv.1803.09010 · Cited by 2,474
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., III, H. D., & Crawford, K. (2021) · Communications of the ACM · doi:10.1145/3458723 · Cited by 2,469
Hateful Symbols or Hateful People? Predictive Features for Hate Speech Detection on Twitter
Waseem, Z., & Hovy, D. (2016) · Proceedings of the NAACL Student Research Workshop · doi:10.18653/v1/n16-2013 · Cited by 1,655 · cited 4×
“Waseem and Hovy (2016) provide the following complete list of terms used in their initial scrape: MKR, asian drive, feminazi, immigrant, nigger, sjw, WomenAgainstFeminism, blameonenotall, islam terrorism, notallmen, victimcard, victim card, arab terror, gamergate, jsil, racecard, race card. dataset…”
BackgroundMethodInfluential
Cited by
Items 401–500 of 865
HOMO-MEX: A Mexican Spanish Annotated Corpus for LGBT+phobia Detection on Twitter
Vásquez, J., Andersen, S., Bel-Enguix, G., Gómez-Adorno, H., & Ojeda-Trueba, S.-L. (2023) · doi:10.18653/v1/2023.woah-1.20 · Cited by 23
Doğruöz, A. S., Sitaram, S., & Yong, Z. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.382 · Cited by 30
Use of Formal Ethical Reviews in NLP Literature: Historical Trends and Current Practices
Santy, S., Rani, A., & Choudhury, M. (2021) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2021.findings-acl.414 · Cited by 9
FairShades: Fairness Auditing via Explainability in Abusive Language Detection Systems
Manerba, M. M., & Guidotti, R. (2021) · IEEE International Conference on Cognitive Machine Intelligence (CogMI) · doi:10.1109/cogmi52975.2021.00014 · Cited by 15
Ethical Awareness in Paralinguistics: A Taxonomy of Applications
Batliner, A. M., Neumann, M., Burkhardt, F., Baird, A., Meyer, S., Vu, N. T., & Schuller, B. W. (2022) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2022.2140385 · Cited by 14
“(Bender & Friedman, 2018).”
Towards Ethics by Design in Online Abusive Content DetectionPreprint
Kiritchenko, S., & Nejadgholi, I. (2020) · arXiv (Cornell University) · doi:10.4224/40002027 · Cited by 18
AI Ethics and Governance in Practice: An IntroductionPreprint
Leslie, D., Rincón, C., Briggs, M., Perini, A., Jayadeva, S., Borda, A., Bennett, S., Burr, C., Aitken, M., Katell, M. A., et al. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4731635 · Cited by 15
Khanjani, Z., Davis, L. S., Tuz, A., Nwosu, K., Mallinson, C., & Janeja, V. P. (2023) · IEEE International Conference on Intelligence and Security Informatics (ISI) · doi:10.1109/isi58743.2023.10297267 · Cited by 25
“[2].”
Using Model Cards for ethical reflection on machine learning models: an interview-based study
Nunes, J. L., Barbosa, G. D. J., de Souza, C. S., & Barbosa, S. D. J. (2024) · Journal on Interactive Systems · doi:10.5753/jis.2024.3444 · Cited by 16
“They focus on different parts of the ML development process: on trained models [Mitchell et al., 2019]; on datasets [Bender and Friedman, 2018; Gebru et al., 2018; Holland et al., 2018]; or even on the development process itself [Hutchinson et al., 2021].”
BackgroundMethodNatural Language Processing in Chatbots: A Review
Patra, B., & Kumar, M. M. (2020) · Türk bilgisayar ve matematik eğitimi dergisi · doi:10.61841/turcomat.v11i3.14655 · Cited by 9
“Issues such as bias in training data, privacy concerns, and the potential misuse of NLP systems must be carefully addressed (Bender & Friedman, 2018).”
BackgroundBabu, C. V. S., Anniyappa, C. S. A., & Dharma Sastha, B. (2024) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3693-3860-5.ch008 · Cited by 8
Magaña, M. I., & Shilton, K. (2025) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3715275.3732148 · Cited by 15
“…of generative models can be opaque and difficult to explain [15], and because the data used to train machine learning models is frequently rife with human biases [3] or an imperfect indicator for unobservable constructs [32], today’s AI systems are frequently untrusted by experts or users [1].”
Morphological Segmentation for Seneca
Liu, Z., Jimerson, R., & Prud’hommeaux, E. T. (2021) · Workshop on Natural Language Processing for Indigenous Languages of the Americas (AmericasNLP) · doi:10.18653/v1/2021.americasnlp-1.10 · Cited by 9
“Following recently advocated scientific practices (Bender and Friedman, 2018; Gebru et al., 2018), we would like to first introduce the data of the indigenous languages to be explored.”
MethodEvaluation Paradigms in Question Answering
Rodríguez, P., & Boyd-Graber, J. L. (2021) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2021.emnlp-main.758 · Cited by 21
“opers of datasets should make their goals clear from the outset (Bender and Friedman, 2018; Gebru et al., 2018).”
BackgroundRicher Countries and Richer Representations
Zhou, K., Ethayarajh, K., & Jurafsky, D. (2022) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2022.findings-acl.164 · Cited by 25
Mind the data gap(s): Investigating power in speech and language datasets
Markl, N. (2022) · Workshop on Language Technology for Equality, Diversity, Inclusion (LT-EDI) · doi:10.18653/v1/2022.ltedi-1.1 · Cited by 19
“Gebru et al. (2021) provide a highly influential documentation framework which can be applied to all AI datasets and Bender and Friedman (2018) introduce an approach to documentation specific to datasets for natural language processing, which I draw on here.”
BackgroundInfluentialTowards a Semantic Approach for Linked Dataspace, Model and Data Cards
Donald, A., Galanopoulos, A., Curry, E., Muñoz, E., Ullah, I., Waskow, M. A., Dabrowski, M., & Kalra, M. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543873.3587659 · Cited by 17
“In the evolution of this concept of card documentation for explainable AI (XAI) components, there have been several names, from Data Statements to Datasheets, to Dataset Nutrition Labels [4, 13, 14].”
BackgroundPiazzolla, S. A., Savoldi, B., & Bentivogli, L. (2024) · HERMES - Journal of Language and Communication in Business · doi:10.7146/hjlcb.vi63.137553 · Cited by 32
DREAMER: a computational framework to evaluate readiness of datasets for machine learning
Ahangaran, M., Zhu, H., Li, R., Yin, L., Jang, J., Chaudhry, A. P., Farrer, L. A., Au, R., & Kolachalama, V. B. (2024) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-024-02544-w · Cited by 10
“…tools have been proposed in prior literature to assess data readiness, including but not limited to AutoML [8], Datasheets [9], Data Statements [10], FactSheets [11], Dataset Nutrition Label [12], Model Cards [13], Data-maid [14], Codebook [15], IBM Data Quality Toolkit [16], and Data…”
MethodBigger Isn’t Better: The Ethical and Scientific Vices of Extra-Large Datasets in Language Models
Goetze, T. S., & Abramson, D. (2021) · ACM Web Science Conference (WebSci) · doi:10.1145/3462741.3466809 · Cited by 14
“A suggestion they make which would apply here is the inclusion of data statements [3].”
BackgroundData-driven Model Generalizability in Crosslinguistic Low-resource Morphological Segmentation
Liu, Z., & Prud’hommeaux, E. T. (2022) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00467 · Cited by 13
“Following recently proposed scientific practices for computational linguistics and NLP (Bender and Friedman, 2018; Gebru et al., 2018), we would like to introduce these languages and their data sets explored in our study (Table 1).”
MethodHavens, L., Alex, B., Bach, B., & Terras, M. (2022) · Workshop on Gender Bias in Natural Language Processing (GeBNLP) · doi:10.18653/v1/2022.gebnlp-1.4 · Cited by 19
“Hutchinson et al. (2021), Mitchell et al. (2019), and Bender and Friedman (2018) propose new documentation methods to facilitate critical interrogation of data and the models trained on them.”
BackgroundA Needle in a Haystack: An Analysis of High-Agreement Workers on MTurk for Summarization
Zhang, L., Mille, S., Hou, Y., Deutsch, D., Clark, E., Liu, Y., Mahamood, S., Gehrmann, S., Clinciu, M., Chandu, K. R., et al. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-long.835 · Cited by 27
Datasets: A Community Library for Natural Language ProcessingPreprint
Lhoest, Q., del Moral, A. V., Jernite, Y., Thakur, A., von Platen, P., Patil, S. S., Chaumond, J., Drame, M., Plu, J., Tunstall, L., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.02846 · Cited by 210
“Datasets is backed by the Dataset Hub 4 that helps users navigate the growing number of available resources and draws inspiration from recent work calling for better documentation of ML datasets in general (Gebru et al., 2018) and NLP datasets in particular (Bender and Friedman, 2018).”
Deciphering the Enigma: A Deep Dive into Understanding and Interpreting LLM OutputsPreprint
Wang, Y. (2023) · doi:10.36227/techrxiv.24085833.v1 · Cited by 9
DOSA: A Dataset of Social Artifacts from Different Indian Geographical Subcultures
Seth, A., Ahuja, S., Bali, K., & Sitaram, S. (2024) · doi:10.2139/ssrn.4756716 · Cited by 8
On the Readiness of Scientific Data Papers for a Fair and Transparent Use in Machine Learning
Giner-Miguelez, J., Gómez, A., & Cabot, J. (2025) · Scientific Data · doi:10.1038/s41597-025-04402-4 · Cited by 14
The 2024 Foundation Model Transparency Index
Bommasani, R., Klyman, K., Kapoor, S., Longpre, S., Xiong, B., Maslej, N., & Liang, P. (2024) · arXiv · doi:10.48550/arxiv.2407.12929 · Cited by 8
Gebru, T., & Denton, E. (2024) · Foundations and Trends® in Computer Graphics and Vision · doi:10.1561/0600000102 · Cited by 16
Trustworthy human computation: a survey
Kashima, H., Oyama, S., Arai, H., & Mori, J. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10974-1 · Cited by 11
“There is a risk of privacy leakage in the explanation of data creation, for example, by disclosing detailed profiles of a small number of crowd workers [22].”
InfluentialNielsen, A., Skylaki, S., Norkutė, M., & Stremitzer, A. (2024) · Journal of Empirical Legal Studies · doi:10.1111/jels.12396 · Cited by 30
The Making of Performative Accuracy in AI Training: Precision Labor and Its Consequences
Zhang, B. Z., Yang, T., Miceli, M., Haimson, O. L., & Thomas, M. (2025) · Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713112 · Cited by 7
“Previous work has underscored the importance of systematically documenting not only datasets but also data production contexts [4, 17, 24, 32, 34].”
RiskRAG: A Data-Driven Solution for Improved AI Model Risk Reporting
Rao, P. S. B., Šćepanović, S., Zhou, K., Bogucka, E. P., & Quercia, D. (2025) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713979 · Cited by 24
“…practices: from model documentation (e.g., model cards [48]), dataset documentation (e.g., datasheets for datasets [24], data statements for NLP [3]), documentation for the purpose of us-ing AI (i.e., ethics sheets for AI tasks [49]), to recent documentation for AI risks (i.e., Risk Cards [17]).”
Review of recent emotion-annotated text corpora and resources
Koufakou, A., & Nieves, E. (2025) · Language Resources and Evaluation · doi:10.1007/s10579-025-09828-1 · Cited by 19
“Data statements have been proposed by Bender and Friedman (2018) to provide guidelines for data curation and reporting.”
Authorship in the age of artificial intelligence
Hajibayova, L., & Lee, S. (2025) · Journal of Documentation · doi:10.1108/jd-04-2025-0109 · Cited by 11
Impact Assessment Card: Communicating Risks and Benefits of AI Uses
Bogucka, E. P., Constantinides, M., Šćepanović, S., & Quercia, D. (2025) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3757482 · Cited by 18
Winslow, B. D., Garibay, Ö. Ö., Goyal, T., Koon, S., Margetis, G., Salvendy, G., Shneiderman, B., Tayebi, A., & Vardoulakis, L. (2026) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2026.2641703 · Cited by 16
Exploring the Intersection of AI, Language, and Law: A Bibliometric Analysis
Lan, Z., & Zhang, X. (2026) · SAGE Open · doi:10.1177/21582440261431000 · Cited by 9
“…model cards with disaggregated performance reporting across jurisdictions and languages, and end-to-end internal algorithmic auditing that links design choices to harm hypotheses and remediation plans (Bender & Friedman, 2018; Gebru et al., 2021; Mitchell et al., 2019; Raji et al., 2020).”
SocioFillmore: A Tool for Discovering Perspectives
Minnema, G., Gemelli, S., Zanchi, C., Caselli, T., & Nissim, M. (2022) · Annual Meeting of the Association for Computational Linguistics: System Demonstrations (ACL Demo) · doi:10.18653/v1/2022.acl-demo.24 · Cited by 7
Dellanzo, A., Cotik, V., Barriga, D. Y. L., Apaza, J. J. M., Palomino, D., Schiaffino, F., Aliaga, A. Y., & Ochoa-Luna, J. E. (2022) · BMC Bioinformatics · doi:10.1186/s12859-022-05094-y · Cited by 7
“In this subsection, we present the process of creation of the corpus, including ethic and data statements [21] and an analysis of the resulting annotated dataset.”
BackgroundMethodArtificial Intelligence, Machine Learning, and Natural Language Processing
Shoenbill, K. A., Kasthurirathne, S. N., & Mendonça, E. A. (2023) · Springer eBooks · doi:10.1007/978-3-031-29171-5_36 · Cited by 14
Designing with Data, Data-Enabled and Data-Driven Design
Funk, M., Lövei, P., & Noortman, R. (2024) · Springer eBooks · doi:10.1007/978-3-319-27648-9_40-1 · Cited by 9
Data Ethics Emergency Drill: A Toolbox for Discussing Responsible AI for Industry Teams
Hanschke, V. A., Rees, D., Alanyali, M., Hopkinson, D., & Marshall, P. (2024) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642402 · Cited by 13
“There have also been proposals of design methods, for example using value sensitive design [21] to support the creation of responsible AI systems during the design phase [6, 22, 55].”
Racial/Ethnic Categories in AI and Algorithmic Fairness: Why They Matter and What They Represent
Mickel, J. (2024) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3630106.3659050 · Cited by 21
“This framework allows for an improved understanding of the assumptions andchoices madeby the usersand developers of datasets and models, helping future dataset and model users understand whether the racial categories are relevant to their use case.”
BackgroundDetecting Independent Pronoun Bias with Partially-Synthetic Data Generation
Munro, R., & Morrison, A. (. (2020) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.157 · Cited by 16
Search with Discretion: Value Sensitive Design of Training Data for Information Retrieval
Iqbal, M., Shilton, K., Sayed, M. F., Oard, D. W., Rivera, J. L., & Cox, W. (2021) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3449207 · Cited by 11
LMSOC: An Approach for Socially Sensitive Pretraining
Kulkarni, V., Mishra, S., & Haghighi, A. (2021) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2021.findings-emnlp.254 · Cited by 19
Text Analysis in Python for Social Scientists: Prediction and Classification
Hovy, D. (2022) · Cambridge University Press eBooks · doi:10.1017/9781108960885 · Cited by 21
Multi-lifespan Envisioning Cards:
Yoo, D., Logler, N., Ballard, S. A., & Friedman, B. (2022) · Designing Interactive Systems Conference · doi:10.1145/3532106.3533495 · Cited by 7
Self-citationVisual Navigation Based on Language Assistance and Memory
Xiao, S., & Fu, W. (2023) · IEEE Access · doi:10.1109/access.2023.3239837 · Cited by 7
Abercrombie, G., Hovy, D., & Prabhakaran, V. (2023) · Linguistic Annotation Workshop · doi:10.18653/v1/2023.law-1.10 · Cited by 31
Markl, N., & Lai, C. (2023) · Interspeech (USB) · doi:10.21437/interspeech.2023-1847 · Cited by 15
“One option could be to include standardised lan-4426 guage variety tags (e.g., BCP-47) as recommended for natural language processing [30].”
BackgroundDiatopIt: A Corpus of Social Media Posts for the Study of Diatopic Language Variation in Italy
Ramponi, A., & Casula, C. (2023) · Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial) · doi:10.18653/v1/2023.vardial-1.19 · Cited by 7
“Data statements (Bender and Friedman, 2018) for D IATOP I T are presented in Appendix A.”
MethodNeural Methods for Data-to-text Generation
Sharma, M., Gogineni, A. K., & Ramakrishnan, N. (2024) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3660639 · Cited by 31
Joint Rumour Stance and Veracity
Lillie, A. E., Middelboe, E. R., & Derczynski, L. (2019) · Conference for Truth and Trust Online (TTO) · doi:10.36370/tto.2019.17 · Cited by 7
“See appendix for data statement (Bender and Friedman, 2018).”
EUREKA: EUphemism Recognition Enhanced through Knn-based methods and Augmentation
Keh, S. S., Bharadwaj, R., Liu, E., Tedeschi, S., Gangal, V., & Navigli, R. (2022) · Workshop on Figurative Language Processing (FigLang) · doi:10.18653/v1/2022.flp-1.15 · Cited by 14
An Information Retrieval Approach to Building Datasets for Hate Speech DetectionPreprint
Rahman, M. M., Balakrishnan, D., Murthy, D., Kutlu, M., & Lease, M. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.09775 · Cited by 23
“For example, harm could come from deploying a machine learning model without considering, mitigating, and/or documenting [37, 12, 15] limitations of its underlying training data, such as the risk of racial bias in hate speech annotations [97, 117, 22].”
BackgroundSkillSpan: Hard and Soft Skill Extraction from English Job PostingsPreprint
Zhang, M., Jensen, K. N., Sonniks, S. D., & Plank, B. (2022) · IT University Of Copenhagen (IT University of Copenhagen) · doi:10.48550/arxiv.2204.12811 · Cited by 30
“eu/esco/skill/A (5)Our data statement (Bender and Friedman, 2018) can be found in Appendix A.”
BackgroundWhat Can Natural Language Processing Do for Peer Review?
Kuznetsov, I., Afzal, O. M., Dercksen, K., Dycke, N., Goldberg, A., Hope, T., Hovy, D., Kummerfeld, J. K., Lauscher, A., Leyton-Brown, K., et al. (2024) · arXiv · doi:10.48550/arxiv.2405.06563 · Cited by 6
“These questions invite additional reflection and consideration while designing a data collection campaign; for a broader discussion see (Bender and Friedman, 2018; Kamocki and Witt, 2020; Rogers et al., 2021).”
BackgroundBlockchain Technology and Cybersecurity in Fintech: Opportunities and Vulnerabilities
Ajayi, O. O., Alozie, C. E., Abieba, O. A., Akerele, J. I., & Collins, A. (2025) · International Journal of Scientific Research in Computer Science Engineering and Information Technology · doi:10.32628/cseit25111210 · Cited by 26
Resource Asymmetry in Multilingual NLP: A Comprehensive Review and Critique
Akindotuni, D. (2025) · Journal of Computer and Communications · doi:10.4236/jcc.2025.137002 · Cited by 12
From tools to co-learners: entangled humanism and the co-evolution of intelligence in AI education
Jose, B., Verghis, A. M., Varghese, S. M., Mumthas, S., & Cherian, J. (2025) · Frontiers in Education · doi:10.3389/feduc.2025.1632990 · Cited by 20
Stoll, A., Yu, J., Andrich, A., & Domahidi, E. (2025) · Communication Methods and Measures · doi:10.1080/19312458.2025.2551693 · Cited by 9
“Such bias in perception can likely be reproduced and forwarded by systems that learn based on human decisions (Bender & Friedman, 2018).”
Model Cards Revisited: Bridging the Gap Between Theory and Practice for Ethical AI Requirements
Puhlfürß, T., Butzke, J., & Maalej, W. (2025) · IEEE International Requirements Engineering Conference (RE) · doi:10.1109/re63999.2025.00034 · Cited by 12
“Data-centric frameworks include data statements [28], dataset nutrition labels [27], data cards [26], and datasheets [25], some of which, unlike the model card framework, pose more explicit ethical questions to the documentation authors.”
Machine learning meets governance: AI innovation decodes corporate violations
Song, F., Huang, Y., & Liu, C. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104539 · Cited by 8
Evolving from Rules to Learning in Urban Modeling and Planning Support Systems
Cai, Z. (2025) · Urban Science · doi:10.3390/urbansci9120508 · Cited by 9
Azhar, T., Mahmud, T., Hasan, M. A., Uddin, M. N., & Park, S.-B. (2026) · Discover Computing · doi:10.1007/s10791-025-09804-x · Cited by 7
“HS identification varies from person to person depending on their age, religion, social status [49].”
Large Language Models as SocioTechnical Systems
Dhole, K. (2023) · doi:10.18653/v1/2023.bigpicture-1.6 · Cited by 25
“…statements and model cards and ii) through pre-prompting (or providing instruction) Documenting datasets and the training data (Gebru et al., 2021; Bender and Friedman, 2018; Stoyanovich and Howe, 2019; Papakyriakopoulos et al., 2023) used could be at least the bare minimum heterogeneous…”
BackgroundThe Iron(ic) Melting Pot: Reviewing Human Evaluation in Humour, Irony and Sarcasm Generation
Loakman, T., Maladry, A., & Lin, C. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.444 · Cited by 23
Human-Centered AI Goals for Speech Therapy Tools
Deka, C., Shrivastava, A., Nautiyal, S., & Chauhan, P. (2023) · Communications in computer and information science · doi:10.1007/978-3-031-49368-3_8 · Cited by 7
Natural Language Processing for HR Chatbots and Virtual Assistants
Singh, A. K., Lakhera, G., Ojha, M., Mishra, A. K., & Kaushik, S. (2024) · Advances in logistics, operations, and management science book series · doi:10.4018/979-8-3693-3318-1.ch020 · Cited by 6
Employing Hybrid AI Systems to Trace and Document Bias in ML Pipelines
Russo, M., Chudasama, Y., Purohit, D., Sawischa, S. F., & Vidal, M.-E. (2024) · IEEE Access · doi:10.1109/access.2024.3427388 · Cited by 11
“Existing methods for the production of documentation and benchmarks for datasets and ML models, such as [12], [18], [19], and [20], propose value-sensitive, human-readable documentation frameworks, while the works of [21] and [22], propose domain-specific and task-oriented, respectively, for…”
MethodSoftware Systems Compliance with the AI Act: Lessons Learned from an International Challenge
Scantamburlo, T., Falcarin, P., Veneri, A., Fabris, A., Gallese, C., Billa, V., Rotolo, F., & Marcuzzi, F. (2024) · doi:10.1145/3643691.3648589 · Cited by 14
The Role of Paradata in Algorithmic Accountability
Trace, C. B., & Hodges, J. A. (2024) · Knowledge management and organizational learning · doi:10.1007/978-3-031-53946-6_11 · Cited by 8
Levine, M. (2025) · British Journal of Social Psychology · doi:10.1111/bjso.12896 · Cited by 6
“…away from just trying to make data bias and inequality transparent, to a more radical path of trying to tackle the underlying relations of power (see for example Bender and Friedman (2018), Buolamwini and Gebru (2018), Mitchell et al. (2019), Gebru (2020), Birhane (2021), Birhane et al. (2022)).”
An Overview of Fairness in Data – Illuminating the Bias in Data Pipeline
Kumar, B. S., Chandrabose, A., & Chakravarthi, B. R. (2021) · Cited by 17
Strathearn, C., & Gkatzia, D. (2022) · DialDoc Workshop on Document-grounded Dialogue and Conversational Question Answering (DialDoc) · doi:10.18653/v1/2022.dialdoc-1.21 · Cited by 10
“Zo Grof !”: A Comprehensive Corpus for Offensive and Abusive Language in Dutch
Ruitenbeek, W., Zwart, V., Noord, R. V. D., Gnezdilov, Z., & Caselli, T. (2022) · Workshop on Online Abuse and Harms (WOAH) · doi:10.18653/v1/2022.woah-1.5 · Cited by 12
“The annotation guidelines, the data, and the code for the reported experiments, and a data statement (Bender and Friedman, 2018) are publicly available.”
MethodInvestigating Debiasing Effects on Classification and Explainability
Manerba, M. M., & Guidotti, R. (2022) · Proceedings of the AAAI/ACM Conference on AI Ethics and Society · doi:10.1145/3514094.3534170 · Cited by 14
“Similar documentation processes applied to data are proposed within Data Statements [3] and Datasheets for Datasets [16].”
MethodEthics consideration sections in natural language processing papers
Benotti, L., & Blackburn, P. R. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.299 · Cited by 11
A Cross-Lingual Study of Homotransphobia on Twitter
Locatelli, D., Damo, G., & Nozza, D. (2023) · Workshop on Cross-Cultural Considerations in NLP (C3NLP) · doi:10.18653/v1/2023.c3nlp-1.3 · Cited by 6
Developing responsible AI practices at the Smithsonian Institution
Dikow, R. B., DiPietro, C., Trizna, M. G., BredenbeckCorp, H., Bursell, M. G., Ekwealor, J. T. B., Hodel, R. G. J., Lopez, N., Mattingly, W., Munro, J., et al. (2023) · Research Ideas and Outcomes · doi:10.3897/rio.9.e113334 · Cited by 11
“Data statements (Bender and Friedman 2018, McMillan-Major et al. 2023), model cards (Mitchell et al. 2019), and datasheets (Gebru et al. 2021) are all focused on better documentation of data and methods, making any work resulting from their use more transparent and reproducible.”
BackgroundGuo, M., Hwa, R., Lin, Y., & Chung, W. (2020) · International Conference on Computational Linguistics · doi:10.18653/v1/2020.coling-main.428 · Cited by 12
Sex and gender bias in natural language processing
Cirillo, D., Gonen, H., Santus, E., Valencia, A., Costa‐jussà, M. R., & Villegas, M. (2022) · Elsevier eBooks · doi:10.1016/b978-0-12-821392-6.00009-1 · Cited by 8
Perez-Almendros, C., Espinosa-Anke, L., & Schockaert, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.08320 · Cited by 6
“An extended data statement (Bender and Friedman, 2018) about the corpus will be published together with the dataset.”
BackgroundHirmer, S. A., Leonard, A., Tumwesige, J., & Conforti, C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2102.02841 · Cited by 6
“A comprehensive dataset characterisation (Bender and Friedman, 2018; Gebru et al., 2018) can help researchers understand whether an existing dataset is appropri-”
BackgroundConvAbuse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational AIPreprint
Curry, A. C., Abercrombie, G., & Rieser, V. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.09483 · Cited by 6
Proceedings of the Eighth International Workshop on Natural Language Processing for Social Media
Kameswari, L., Sravani, D., Mamidi, R., Hiware, K., Dutt, R., Sinha, S., Patro, S., Ghosh, K., Ghosh, S., Helbig, D., et al. (2020) · doi:10.18653/v1/2020.socialnlp-1 · Cited by 2
Nasarian, E., Alizadehsani, R., Acharya, U. R., & Tsui, K. (2023) · Research Square · doi:10.21203/rs.3.rs-3626164/v2 · Cited by 7
Lessons from the Trenches on Reproducible Evaluation of Language Models
Biderman, S., Schoelkopf, H., Sutawika, L., Gao, L., Tow, J., Abbasi, B., Aji, A. F., Ammanamanchi, P. S., Black, S., Clive, J., et al. (2024) · arXiv · doi:10.48550/arxiv.2405.14782 · Cited by 9
“…et al., 2020) which may influence the direction of future research and the adoption of new methods by the community (Dehghani et al., 2021), or lead to adverse effects from deploying suboptimal or harmful models (Bender and Friedman, 2018) on tasks for which they are ill-suited (Raji et al., 2022).”
“I’ll be there for you”: The One with Understanding Indirect Answers
Damgaard, C., Toborek, P., Eriksen, T. N., & Plank, B. (2021) · Workshop on Computational Approaches to Discourse (CODI) · doi:10.18653/v1/2021.codi-main.1 · Cited by 5
Sharma, P., & Dash, B. (2023) · International Journal on Cybernetics & Informatics · doi:10.5121/ijci.2023.120103 · Cited by 7
Browsing and Searching Metadata of TREC
Breuer, T., Voorhees, E. M., & Soboroff, I. M. (2024) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3626772.3657873 · Cited by 6
Bias in Federated Learning: Factors, Effects, Mitigations, and Open Issues
Benmalek, M., & Seddiki, A. (2024) · Ingénierie des systèmes d information · doi:10.18280/isi.290605 · Cited by 13
Responsible guidelines for authorship attribution tasks in NLP
Saxena, V., Tamò‐Larrieux, A., DIJCK, G. V., & Spanakis, G. (2025) · Ethics and Information Technology · doi:10.1007/s10676-025-09821-w · Cited by 14
“While removing personally identifiable information (Dev et al., 2022) and demographic attributes (Bender & Friedman, 2018) helps, it does not necessarily correct the underlying demographic biases that influence model predictions.”
Applying Data Feminism Principles to Assess Bias in English and Arabic NLP Research
Mayeda, C., Singh, A., Mahale, A., Sakr, L. S., & ElSherief, M. (2025) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3715275.3732119 · Cited by 7
“However, English papers typically do not explicitly mention ”English” in their titles, as observed by Bender and Friedman [3].”
Influential“I'm” Lost in Translation: Pronoun Missteps in Crowdsourced Data Sets
Seaborn, K., & Kim, Y. (2023) · CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) · doi:10.1145/3544549.3585667 · Cited by 9
Metadata
- Authors
- Bender, E. M., & Friedman, B.
- Affiliations
- University of Washington (US)
- Corresponding author
- Bender, Emily M.
- Friedman, Batya
- Venue
- Transactions of the Association for Computational Linguistics
- Publisher
- MIT Press
- Published
- 2018-12
- Volume
- 6
- Pages
- 587-604
- ISSN
- 2307-387X (ISSN-L)
- OpenAlex
- W2911227954
- Semantic Scholar
- 97bfa89addc6e5d76361e4c1e296949cad887b86
- Cited by
- 854
- References
- 46
- Citations by year
- 2026: 108 · 2025: 96 · 2024: 102 · 2023: 154 · 2022: 127 · 2021: 177 · 2020: 68 · 2019: 20 · 2018: 1
- FWCI
- 132.62
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- cc-by
- Locations
- Transactions of the Association for Computational Linguistics (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Funders
- National Science Foundation (1302709)
- University of Washington
- York University
- Macquarie University
- University of California, San Diego
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- natural language processing, algorithmic bias, underrepresented groups, ethical practice, linguistic diversity
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
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.