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 501–600 of 865
It Takes Two to Tango: Navigating Conceptualizations of NLP Tasks and Measurements of Performance
Subramonian, A., Yuan, X., Daumé, H., & Blodgett, S. L. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.202 · Cited by 45
Madureira, B., & Schlangen, D. (2023) · Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2023.eacl-main.169 · Cited by 24
Social Commonsense for Explanation and Cultural Bias Discovery
Bauer, L., Tischer, H., & Bansal, M. (2023) · Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2023.eacl-main.271 · Cited by 15
“By making explicit the underlying cultural assumptions and causal relationships, we are able to identify the biases that data from a certain source may have, which is paramount when using this data in the development of technology for other applications (Bender and Friedman, 2018).”
K-HATERS: A Hate Speech Detection Corpus in Korean with Target-Specific Ratings
Park, C., Kim, S., Park, K., & Park, K. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.952 · Cited by 13
Zou, L. (2024) · Applied Mathematics and Nonlinear Sciences · doi:10.2478/amns-2024-2842 · Cited by 5
Towards Certifying Trustworthy Machine Learning Systems
Yap, R. H. C. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-73959-1_7 · Cited by 8
Automatic Resolution of Domain Name Disputes
Vihikan, W. O., Mistica, M., Levy, I., Christie, A. F., & Baldwin, T. (2021) · Natural Legal Language Processing Workshop (NLLP) · doi:10.18653/v1/2021.nllp-1.24 · Cited by 5
leondz/Political-Stance-in-Danish: v2
Lehmann, R., & Derczynski, L. (2020) · DSpace repository (University of Tartu) · doi:10.5281/zenodo.3862148 · Cited by 5
2kenize: Tying Subword Sequences for Chinese Script Conversion
Pranav, A., & Augenstein, I. (2020) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2020.acl-main.648 · Cited by 8
Cross-Lingual Cross-Domain Nested Named Entity Evaluation on English Web Texts
Plank, B. (2021) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2021.findings-acl.158 · Cited by 5
Frequency-based Distortions in Contextualized Word EmbeddingsPreprint
Zhou, K., Ethayarajh, K., & Jurafsky, D. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.08465 · Cited by 19
“As previous literature has suggested (Gebru et al., 2018; Mitchell et al., 2019; Bender and Friedman, 2018), documentation for datasets is critical for increasing transparency and accountability in machine learning models.”
Detection of Machine-Generated Text: Literature Survey
Valiaiev, D. (2024) · arXiv · doi:10.48550/arxiv.2402.01642 · Cited by 5
Cooper, N., Heldreth, C., & Hutchinson, B. (2024) · Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/2024.eacl-short.19 · Cited by 5
On the State of Reporting in Crowdsourcing Experiments and a Checklist to Aid Current Practices
Ramírez, J., Sayin, B., Baez, M., Casati, F., Cernuzzi, L., Benatallah, B., & Demartini, G. (2021) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3479531 · Cited by 4
“Examples of this can be found in evidence-based medicine where systematic reviews follow a strict elaboration protocol [71], or, more recently, in the machine learning community where authors are encouraged to follow pre-established checklists or datasheets to communicate their models and datasets effectively [2, 6, 2…”
BackgroundNLPositionality: Characterizing Design Biases of Datasets and Models
Santy, S., Liang, J. T., Bras, R. L., Reinecke, K., & Sap, M. (2023) · arXiv · doi:10.48550/arxiv.2306.01943 · Cited by 31
“The challenge is that design biases arise from the myriad of design choices made; in the context of creating datasets and models, only some of these choices may be documented (e.g., through model cards and data sheets; Bender and Friedman, 2018; Mitchell et al., 2019; Gebru et al., 2021).”
BackgroundFrom Policy to Pipeline: A Governance Framework for AI Development and Operations Pipelines
Butt, T. A., Iqbal, M., & Arshad, N. (2025) · IEEE Access · doi:10.1109/access.2025.3647479 · Cited by 14
“Additional refinements such as Data Statements [25] and Dataset Nutrition Labels [26] increase granularity and domain specificity.”
Dialogic Reflection and Algorithmic Bias: Pathways Toward Inclusive AI in Education
García, P. P., Aza, M. J. D., & Feltrero, R. (2026) · Trends in Higher Education · doi:10.3390/higheredu5010009 · Cited by 19
“As Bender & Friedman [47] point out, interpreting algorithmic biases is not limited to identifying technical errors but requires recognizing how data and AI systems reproduce sociocultural inequalities in educational environments.”
Khan, H., AlQahtani, F., Jiang, W., Ali, M. H., Amin, S. U., & Koo, J. (2026) · Computer Science Review · doi:10.1016/j.cosrev.2026.101037 · Cited by 5
Harvey, D., Rayson, P., Lobban, F. A., Palmier‐Claus, J. E., Dolman, C., Chataigné, A., & Jones, S. H. (2025) · JMIR Infodemiology · doi:10.2196/65632 · Cited by 6
“Reporting on aspects such as gender also contributes toward more ethical natural language processing data collection as these predictions can suggest how experimental results might be generalized and also highlights where the data include bias [78].”
Add-Remove-or-Relabel: Practitioner-Friendly Bias Mitigation via Influential Fairness
Richardson, B., Sattigeri, P., Wei, D., Ramamurthy, K. N., Varshney, K. R., Dhurandhar, A., & Gilbert, J. E. (2023) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3593013.3594039 · Cited by 14
“main, such as natural language processing [13, 15, 18] and computer vision [56, 108, 113].”
BackgroundMapping the Smart City: Participatory approaches to XAI
Gupta, S. (2023) · Designing Interactive Systems Conference · doi:10.1145/3563703.3593063 · Cited by 7
“Several methods and tools, such as model [44] and data documentation [4, 7, 22, 29] frameworks, have been designed to explain AI to experts [17].”
MethodSahoo, N., Mallela, N., & Bhattacharyya, P. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.842 · Cited by 16
The State of Profanity Obfuscation in Natural Language Processing Scientific Publications
Nozza, D., & Hovy, D. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.240 · Cited by 18
Is NLP Ready for Standardization?
Aufrant, L. (2022) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2022.findings-emnlp.202 · Cited by 7
“This is another area where individual initiatives have produced valuable guidelines on necessary meta-data (Bender and Friedman, 2018).”
Unifying Data Perspectivism and Personalization: An Application to Social Norms
Plepi, J., Neuendorf, B., Flek, L., & WELCH, C. A. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.500 · Cited by 44
Xu, S., Staufer, L., T.y.s.s, S., Ichim, O., Heri, C., & Grabmair, M. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.718 · Cited by 16
Owate, C. N., & David-West, B. (2024) · American Journal of Education and Information Technology · doi:10.11648/j.ajeit.20240801.17 · Cited by 7
CommitBench: A Benchmark for Commit Message Generation
Schall, M., Czinczoll, T., & de Melo, G. (2024) · IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER) · doi:10.1109/saner60148.2024.00080 · Cited by 19
“Another important aspect is that the growing need for reproducibility and extensive meta-data for training and evaluation datasets [36], [37].”
BackgroundThe Data Artifacts Glossary: a community-based repository for bias on health datasets
Gameiro, R. R., Woite, N. L., Sauer, C. M., Hao, S., Fernandes, C. O., Premo, A. E., Teixeira, A. R., Resli, I., Wong, A.-K. I., & Celi, L. A. (2025) · Journal of Biomedical Science · doi:10.1186/s12929-024-01106-6 · Cited by 7
“Notable among these efforts are Data Cards [19], Data Statements [20], Datasheet for Datasets [21], Model Cards [22], AI-Usage Cards [23], and the Dataset Nutritional Label [18].”
Network Report: A Structured Description for Network Datasets
Zheng, X., Rossi, R. A., Ahmed, N. K., & Moritz, D. (2022) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3511808.3557115 · Cited by 6
“dataset documentation and intended uses accompanying the main paper, and recommends authors to follow existing frameworks such as Datasheets [17], Data Statements for NLP [4], and Nutrition La-”
BackgroundMethodInfluentialCTM - A Model for Large-Scale Multi-View Tweet Topic Classification
Kulkarni, V., Leung, K., & Haghighi, A. (2022) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/2022.naacl-industry.28 · Cited by 4
Rawlsian AI fairness loopholes
Jørgensen, A. K., & Søgaard, A. (2022) · AI and Ethics · doi:10.1007/s43681-022-00226-9 · Cited by 5
Ethical and Legal Challenges of Digital Medicine in Pandemics
Minssen, T., & Gerke, S. (2023) · Springer eBooks · doi:10.1007/978-3-662-66872-6_12 · Cited by 5
Sekara, V., Karsai, M., Moro, E., Kim, D., Delamónica, E., Cebrián, M., Luengo-Oroz, M. A., Jiménez, R. M., & García–Herranz, M. (2024) · Advances in Complex Systems · doi:10.48550/arxiv.2307.01891 · Cited by 12
Alshahrani, S., Wali, E., & Matthews, J. N. (2022) · Arabic Natural Language Processing Workshop (WANLP) · doi:10.18653/v1/2022.wanlp-1.34 · Cited by 11
A Dataset of Sustainable Diet Arguments on Twitter
Hansen, M., & Hershcovich, D. (2022) · Workshop on NLP for Positive Impact (NLP4PI) · doi:10.18653/v1/2022.nlp4pi-1.5 · Cited by 8
Our kind of people? Detecting populist references in political debates
Klamm, C., Rehbein, I., & Ponzetto, S. P. (2023) · Findings of the Association for Computational Linguistics: EACL · doi:10.18653/v1/2023.findings-eacl.91 · Cited by 5
Trigger Warnings: Bootstrapping a Violence Detector for Fan Fiction
Wolska, M. A., Wiegmann, M., Schröder, C., Borchardt, O., Stein, B., & Potthast, M. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.41 · Cited by 9
Bornholmsk Natural Language Processing: Resources and Tools.
Derczynski, L., & Kjeldsen, A. S. (2019) · DSpace repository (University of Tartu) · Cited by 4
An IDR Framework of Opportunities and Barriers between HCI and NLP
Inie, N., & Derczynski, L. (2021) · IT University Of Copenhagen (IT University of Copenhagen) · Cited by 9
Identifying communicative functions in discourse with content types
Caselli, T., Rachele, S., & Moretti, G. (2021) · Language Resources and Evaluation · doi:10.1007/s10579-021-09554-4 · Cited by 9
Abusive Language Recognition in Russian
Saitov, K., & Derczynski, L. (2021) · IT University Of Copenhagen (IT University of Copenhagen) · Cited by 4
Efficiently Mitigating Classification Bias via Transfer Learning.Preprint
Jin, X., Barbieri, F., Davani, A. M., Kennedy, B., Neves, L., & Ren, X. (2020) · arXiv (Cornell University) · Cited by 5
Hate Speech Classifiers Learn Human-Like Social StereotypesPreprint
Davani, A. M., Atari, M., Kennedy, B., & Dehghani, M. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.14839 · Cited by 4
“Annotated datasets of hate speech rarely report psychological assessments of their annotators (Bender & Friedman, 2018; Prabhakaran et al., 2021), and even if they do, little variance may exist among the few annotators who code an item.”
BackgroundDon’t 'Research Fast and Break Things': On the Ethics of Computational Social SciencePreprint
Leslie, D. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4134621 · Cited by 4
Editorial: Bias, Subjectivity and Perspectives in Natural Language Processing
Basile, V., Caselli, T., Balahur, A., & Ku, L. (2022) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2022.926435 · Cited by 9
“…models) is growing in the NLP community with dedicated efforts to capture, remove, and understand the effect of bias in the data and on the prediction capabilities of the models ( Bolukbasi et al., 2016; Bender and Friedman, 2018; Gonen and Goldberg, 2019; Bartl et al., 2020; Bender et al., 2021).”
BackgroundLim, S., & Pérez‐Ortiz, M. (2024) · Research Square · doi:10.21203/rs.3.rs-5283007/v1 · Cited by 12
Critical, but constructive: defining, detecting, and addressing bias in Computational Social Science
Hase, V., Bachl, M., & TeBlunthuis, N. (2025) · Communication Methods and Measures · doi:10.1080/19312458.2025.2575468 · Cited by 4
“Technical fields have pioneered, for example, model cards for machine learning classifiers (Mitchell et al., 2019) and data statements in NLP (Bender & Friedman, 2018), though these also gain traction in the social sciences (Kostygina et al., 2023).”
A Semantic Approach for Linked Model, Data, and Dataspace Cards
Donald, A., Galanopoulos, A., Curry, E., Muñoz, E., Ullah, I., Waskow, M. A., Kalra, M., Saxena, S., & Iqbal, T. (2025) · IEEE Access · doi:10.1109/access.2025.3572211 · Cited by 3
“Since, in the modern world, we can collect and utilise large-scale data for the solution of different complex applications, it is of utmost importance to understand and use this data responsibly to mitigate potential risks and critical applications [12], [13], [14].”
InfluentialWhose news? Critical methods for assessing bias in large historical datasets
Beelen, K., Lawrence, J., McDonough, K., & Wilson, D. C. S. (2025) · Computational Humanities Research · doi:10.1017/chr.2025.10007 · Cited by 3
“…and cultural limitations could lead to emergent bias, usually described as a demographic mismatch between the source (a model’s training data) and target (users of this model) (Hovy and Prabhumoye 2021; Bender and Friedman 2018), which may affect humanities research using digi-tised collections.”
The Software Diversity Card: A framework for reporting diversity in software projects
Giner-Miguelez, J., Morales, S., Cobos, S., Izquierdo, J. L., Clarisó, R., & Cabot, J. (2025) · Information and Software Technology · doi:10.1016/j.infsof.2025.107950 · Cited by 5
What is a Data Document? Analyzing Four Emerging Data Documentation Frameworks in AI / ML
Maemura, E. (2025) · Proceedings of the Association for Information Science and Technology · doi:10.1002/pra2.1270 · Cited by 3
“Origins, Purpose and Goals Data Statements for Natural Language Processing (NLP) originate in a proposal by Bender and Friedman (2018), researchers at the University of Washington.”
InfluentialMontero, A. M. B., Huet‐Dastarac, M., Herranz-Hernández, S. M., Tengler, B., Buhl, E. S., Galapon, A., Cárdenas, C. E. S., Fusella, M., Herbin, G., de Hond, Y. J. M., et al. (2026) · Physics and Imaging in Radiation Oncology · doi:10.1016/j.phro.2026.100940 · Cited by 7
Erlei, A., Cau, F. M., Georgiev, R., Kumar, S. C., Bizer, K., & Gadiraju, U. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791420 · Cited by 7
“…training data quality are powerful and simple signals for lay users when considering AI models (e.g., similar to model cards or AI nutrition labels [13, 55, 80, 103, 128, 136]), future work may expand on our results by deploying “real” AI systems and introducing higher-dimensional quality…”
Kapwata, T., Webster, C., Chen, W. G. G., Suder, M. I., Jarvis, D., Vogt, A., Mahlangeni, N. T., Laban, T. L., Naidoo, N., Mudu, P., et al. (2026) · Environmental Monitoring and Assessment · doi:10.1007/s10661-026-15376-0 · Cited by 5
“A primary weakness of ML models is their need for data and sensitivity to biases in the data (Bender & Friedman, 2018; Bender et al., 2021; Brown et al., 2024; Walter, 2024).”
Hochgesang, J. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009734349 · Cited by 4
Algorithmic Status Inequality: An Integrative Perspective on AI‐Driven Social Stratification
Wu, J. (2026) · Journal of Management Studies · doi:10.1111/joms.70134 · Cited by 6
Srinivasan, R., Denton, E., Famularo, J., Rostamzadeh, N., Díaz, F., & Coleman, B. (2021) · neural information processing systems · Cited by 25
DataDoc Analyzer: A Tool for Analyzing the Documentation of Scientific Datasets
Giner-Miguelez, J., Gómez, A., & Cabot, J. (2023) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3583780.3614737 · Cited by 11
Directions of Technical Innovation for Regulatable AI Systems
Shen, X., Brown, H., Tao, J., Strobel, M., Tong, Y., Narayan, A., Soh, H., & Doshi-Velez, F. (2024) · Communications of the ACM · doi:10.1145/3653670 · Cited by 3
Developing Linguistic Patterns to Mitigate Inherent Human Bias in Offensive Language Detection
Tanyel, T., Alkurdi, B., & Ayvaz, S. (2024) · TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES · doi:10.55730/1300-0632.4105 · Cited by 4
Intelligent tunnel lining defect detection: advances in image acquisition and data-driven techniques
Yao, Y., Zhao, Y., Li, J., Wang, F., & Liu, C. (2025) · Intelligent Transportation Infrastructure · doi:10.1093/iti/liaf013 · Cited by 4
Park, J., & Jeoung, S. (2022) · doi:10.18653/v1/2022.nlppower-1.1 · Cited by 10
“Recent works have started to question the validity of such benchmark datasets regarding their generalizability (Bowman and Dahl, 2021; Paullada et al., 2021), documentation practices (Bender and Friedman, 2018), and social impact (Hovy and Spruit, 2016; Sap et al., 2021), amongst others.”
BackgroundMaking Smart Cities Explainable: What XAI Can Learn from the “Ghost Map”
Gupta, S., & Loukissas, Y. A. (2023) · CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) · doi:10.1145/3544549.3585847 · Cited by 4
“Alongside the creation of model documentation frameworks, XAI researchers have also designed systems to specifically document data creation processes such as “Data-centric Explanations” [2], “Datasheets” [16], “Dataset Nutrition Labels” [23], and “Data Statements” [4].”
MethodResources for Automated Identification of Online Gender-Based Violence: A Systematic Review
Abercrombie, G., Jiang, A., Gerrard-Abbott, P., Konstas, I., & Rieser, V. (2023) · doi:10.18653/v1/2023.woah-1.17 · Cited by 30
“In documenting resources, authors should provide full data statements or similar (e.g. Bender and Friedman, 2018; Díaz et al., 2022), and, to preserve minority voices, dataset releases should includenon-aggregated labels (Prabhakaran et al., 2021).”
BackgroundRisks and NLP Design: A Case Study on Procedural Document QA
Haduong, N., Gao, A., & Smith, N. A. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.81 · Cited by 6
“…theme is the lack of rigorous assessment or guidelines for deploying models to end users (Tan et al., 2022; Ganguli et al., 2022), with work in mitigating harms operating broadly over large, diverse settings (Blodgett et al., 2020; Buiten, 2019; Zhang et al., 2022; Bender and Friedman, 2018).”
BackgroundNatCS: Eliciting Natural Customer Support Dialogues
Gung, J., Moeng, E., Rose, W., Gupta, A., Zhang, Y., & Mansour, S. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.613 · Cited by 24
Detection of Hate Speech Using Improved Deep Learning Techniques
Gallardo, J. N., Gloria, E. D. G., Landicho, N. R. P., & Sueno, H. T. (2023) · International Conference on Information Technology, Computer, and Electrical Engineering · doi:10.1109/icitacee58587.2023.10277103 · Cited by 3
Business Data Ethics: Emerging Models for Governing AI and Advanced Analytics
Hirsch, D. D., Bartley, T., Chandrasekaran, A., Norris, D., Parthasarathy, S., & Turner, P. N. (2023) · SpringerBriefs in law · doi:10.1007/978-3-031-21491-2 · Cited by 11
Completeness of Datasets Documentation on ML/AI Repositories: An Empirical Investigation
Rondina, M., Vetrò, A., & Martin, J. C. D. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-49008-8_7 · Cited by 9
“The list of Test Fields is largely based on Datasheets for Datasets [8], with some further insights from relevant documentation standardization proposals in the literature [10,3].”
Bias and Fairness in Machine Learning
Faujdar, P., Jain, D., & Verma, V. (2020) · International Journal of Psychosocial Rehabilitation · doi:10.61841/v24i5/400346 · Cited by 3
A Blockchain-Based Approach for Model Card Accountability and Regulatory Compliance
Lohachab, A., & Urovi, V. (2024) · Lecture notes in business information processing · doi:10.1007/978-3-031-61003-5_4 · Cited by 4
What Is Your Favorite Gender, MLM? Gender Bias Evaluation in Multilingual Masked Language Models
Yu, J., Kim, S. U., Choi, J., & Choi, J. D. (2024) · Information · doi:10.3390/info15090549 · Cited by 8
Anik, A. I., & Bunt, A. (2024) · IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC) · doi:10.1109/vl/hcc60511.2024.00024 · Cited by 6
“Given the importance of training datasets in ML and AI systems, there is ongoing work on documenting and presenting training data to those with specific expertise (e.g., ML developers, data specialists, and domain experts) [48], [49], [50], [51], [52], [53], [54].”
BackgroundAI Data Development: A Scorecard for the System Card Framework
Bahiru, T. K., Tibebu, H., & Kakadiaris, I. A. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-8687-2_42 · Cited by 3
Duluth at SemEval-2020 Task 12: Offensive Tweet Identification in English with Logistic Regression
Pedersen, T. (2020) · International Workshop on Semantic Evaluation (SemEval) · doi:10.18653/v1/2020.semeval-1.255 · Cited by 4
Ethical Sensitivity in Machine Learning Development
Boyd, K. (2020) · ACM Conference on Computer Supported Cooperative Work (CSCW) · doi:10.1145/3406865.3418359 · Cited by 4
“Training data has also become a target for accountability interventions, like the Datasheets, the dataset nutrition label, and data statements for NLP [1, 4, 5] and is described by industry practitioners as a key place to intervene to support fairness in ML [6].”
BackgroundGuideline Bias in Wizard-of-Oz Dialogues
Hansen, V. P. B., & Søgaard, A. (2021) · doi:10.18653/v1/2021.bppf-1.2 · Cited by 3
Contextualizing Variation in Text Style Transfer Datasets
Schoch, S., Du, W., & Ji, Y. (2021) · International Conference on Natural Language Generation · doi:10.18653/v1/2021.inlg-1.22 · Cited by 4
Modeling Disclosive Transparency in NLP Application Descriptions
Saxon, M., Levy, S., Wang, X., Albalak, A., & Wang, W. Y. (2021) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2021.emnlp-main.153 · Cited by 8
“Disclosing potential deficiencies, biases, failure points, and intended use cases for datasets (Holland et al., 2018; Bender and Friedman, 2018; Gebru et al., 2018), pretrained models (Mitchell et al.”
MethodDetecting Unassimilated Borrowings in Spanish: An Annotated Corpus and Approaches to Modeling
Mellado, E. Á., & Lignos, C. (2022) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2022.acl-long.268 · Cited by 4
PANCETTA: Phoneme Aware Neural Completion to Elicit Tongue Twisters Automatically
Keh, S. S., Feng, S. Y., Gangal, V., Alikhani, M., & Hovy, E. H. (2023) · Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2023.eacl-main.36 · Cited by 5
Integrating human and machine coding to measure political issues in ethnic newspaper articles
Kim, J. Y. (2021) · Journal of Computational Social Science · doi:10.1007/s42001-020-00097-2 · Cited by 3
“Data annotation’s role in machine learning has been debated among scholars concerning bias and fairness in machine learning [5, 7, 36, 80].”
BackgroundDiscovering the Sweet Spot of Human-Computer Configurations: A Case Study in Information ExtractionPreprint
Mackeprang, M., Müller-Birn, C., & Stauss, M. T. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.07065 · Cited by 2
Easy, Reproducible and Quality-Controlled Data Collection with CrowdaqPreprint
Ning, Q., Wu, H., Dasigi, P., Dua, D., Gardner, M., Logan, R. L., Marasović, A., & Nie, Z. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.06694 · Cited by 3
“Dodge et al. (2019) and Pineau (2020) provide suggestions for system re-producibility, and Bender and Friedman (2018) and Gebru et al. (2018) propose “data statements” and “datasheets for datasets” for data collection reproducibility.”
BackgroundDetecting Abusive AlbanianPreprint
Nurce, E., Keci, J., & Derczynski, L. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2107.13592 · Cited by 11
“This way, as mentioned in Bender and Friedman (2018), the dataset will reflect a wholesome portrait of how Albanian language is used throughout platforms including different dialects, writing styles, content types and tone of language.”
Deciphering the Enigma: A Deep Dive into Understanding and Interpreting LLM OutputsPreprint
Wang, Y. (2023) · doi:10.36227/techrxiv.24085833 · Cited by 6
Text Mining for Bioinformatics Using Biomedical Literature
Lamúrias, A., Sousa, D., & Couto, F. M. (2024) · Elsevier eBooks · doi:10.1016/b978-0-323-95502-7.00017-8 · Cited by 2
Bernhard, D., & Dolińska, J. (2025) · Corpus · doi:10.4000/1364t · Cited by 2
AI Chatbot for Educational Institution: A Case Study
R, D. P., Sanghavi, E., Chennakeshava, G., Kodabagi, M. M., Ali, B., & Gayathri, H. P. (2025) · International Conference on Knowledge Engineering and Communication Systems (ICKECS) · doi:10.1109/ickecs65700.2025.11034867 · Cited by 5
Multilingual Relation Extraction: A Survey
Ali, M., Speck, R., Zahera, H. M., Saleem, M., Moussallem, D., & Ngomo, A.-C. N. (2025) · IEEE Access · doi:10.1109/access.2025.3604258 · Cited by 5
“Additional linguistic features, metadata, and ethical considerations such as privacy, consent, and bias mitigation are essential [101].”
Bauman, T., Fisher, Y., Kizel, F., Omer, I., & Dalyot, S. (2025) · Geo-spatial Information Science · doi:10.1080/10095020.2025.2543498 · Cited by 2
“A rich set of descriptions enables a deeper understanding of the context, terms, and content that are associated with each location (Bender and Friedman 2018), while a diverse set of place types enhances the dataset’s overall coverage, applicability, and effectiveness for TG implementation…”
Asseri, B., Abaker, E., Mogren, M. A., Alhefdhi, T., & Al‐Wabil, A. (2025) · AI · doi:10.3390/ai6090211 · Cited by 6
Toward substantive intersectional algorithmic fairness: desiderata for a feminist approach
Mirsch, M., Wegner, L., Strube, J., & Leicht-Scholten, C. (2026) · AI and Ethics · doi:10.1007/s43681-025-00926-y · Cited by 4
“Practice s such as “datasheets for datasets” [107] , “data statements” [108] , or “model cards” [109] can support this reflexive documentation, but their use should be embedded within participatory design processes that give affected communities the authority to steer, contest, or veto data and…”
Finding our ROLE: How and why to reframe essentialist approaches to language
Namboodiripad, S., Kutlu, E., Babel, A. M., Babel, M., Baese-Berk, M., Bassuk, P. B., Block, A., Pérez, R. C., Carlson, M. T., Carraturo, S., et al. (2026) · Cognition · doi:10.1016/j.cognition.2026.106444 · Cited by 12
Gasparotto, M. (2026) · Big Data & Society · doi:10.1177/20539517251406184 · Cited by 5
“In recent years, a critical mass of researchers in computational linguistics (CL) and allied fi elds have begun speaking out on the problem of bias introduced during the language dataset development and modeling process (Basta et al., 2019; Bender et al., 2021; Bender and Friedman, 2018).”
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.