Quantifying Social Biases in NLP: A Generalization and Empirical Comparison of Extrinsic Fairness Metrics
Open accessThe DOI matches its Crossref record; 83 of the 84 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
Measuring bias is key for better understanding and addressing unfairness in NLP/ML models. This is often done via fairness metrics, which quantify the differences in a model’s behaviour across a range of demographic groups. In this work, we shed more light on the differences and similarities between the fairness metrics used in NLP. First, we unify a broad range of existing metrics under three generalized fairness metrics, revealing the connections between them. Next, we carry out an extensive empirical comparison of existing metrics and demonstrate that the observed differences in bias measurement can be systematically explained via differences in parameter choices for our generalized metrics.
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 check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Most of the existing bias evaluation metrics have been for specific categories of bias, such as gender bias (Czarnowska et al. 2021).”
From Towards trustworthy LLMs: a review on debiasing and dehallucinating in large language models (2024) · cites this paper as Czarnowska et al. 2021
Experimental — accuracy not yet measured
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
On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other
Mann, H. B., & Whitney, D. R. (1947) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177730491 · Cited by 14,320 · cited 1×
“(2019) define their comparison function using the Mann-Whitney U test statistic (Mann and Whitney, 1947).”
MethodConditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
Lafferty, J., McCallum, A., & Pereira, F. C. N. (2001) · ScholarlyCommons (University of Pennsylvania) · Cited by 12,935
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank
Socher, R., Perelygin, A., Wu, J., Chuang, J., Manning, C. D., Ng, A., & Potts, C. (2013) · Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d13-1170 · Cited by 6,878 · cited 2×
“Models We experiment on three RoBERTa (Liu et al., 2019) based models: 6 (i) a binary classifier 6 Our preliminary experiments also used models based on Electra (Clark et al., 2020) as well as those trained on SST-2 and SST-3 datasets (Socher et al., 2013).”
MethodA Survey on Bias and Fairness in Machine Learning
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., & Galstyan, A. (2022) · ACM Computing Surveys · doi:10.1145/3457607 · Cited by 5,812 · cited 1×
“Surveys of Mehrabi et al. (2019), Hutchinson and Mitchell (2019) and Chouldechova and Roth (2020) cover a broad scope of literature on algorithmic fairness.”
BackgroundA Survey on Bias and Fairness in Machine Learning
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., & Galstyan, A. (2019) · arXiv · doi:10.48550/arxiv.1908.09635 · Cited by 5,812
Dwork, C., Hardt, M., Pitassi, T., Reingold, O., & Zemel, R. (2012) · Proceedings of the 3rd Innovations in Theoretical Computer Science Conference · doi:10.1145/2090236.2090255 · Cited by 3,501
Semantics derived automatically from language corpora contain human-like biases
Caliskan, A., Bryson, J. J., & Narayanan, A. (2017) · Science · doi:10.1126/science.aal4230 · Cited by 2,887 · cited 2×
“A related line of research measures bias present in sentence or word representations (Bolukbasi et al., 2016; Caliskan et al., 2017; Kurita et al., 2019; Sedoc and Ungar, 2019; Chaloner and Maldonado, 2019; Dev and Phillips, 2019; Gonen and Goldberg, 2019; Hall Maudslay et al., 2019; Liang et al., 2020; Shin et al., 2…”
BackgroundIntroduction to the CoNLL-2003 Shared Task: Language-Independent Named Entity Recognition
Tjong Kim Sang, E. F., & De Meulder, F. (2003) · Proceedings of the seventh conference on Natural language learning at HLT-NAACL 2003 - · doi:10.3115/1119176.1119195 · Cited by 2,788
Equality of Opportunity in Supervised Learning
Hardt, M., Price, E., & Srebro, N. (2016) · arXiv · doi:10.48550/arxiv.1610.02413 · Cited by 2,536 · cited 2×
“The original versions target demographic parity (Dwork et al., 2012), while the TC versions target equality of opportunity (Hardt et al., 2016) and can pinpoint the existence of bias more precisely, as we show later (Section 6).”
BackgroundMan is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
Bolukbasi, T., Chang, K.-W., Zou, J., Saligrama, V., & Kalai, A. (2016) · arXiv · doi:10.48550/arxiv.1607.06520 · Cited by 1,615 · cited 1×
“A related line of research measures bias present in sentence or word representations (Bolukbasi et al., 2016; Caliskan et al., 2017; Kurita et al., 2019; Sedoc and Ungar, 2019; Chaloner and Maldon-ado, 2019; Dev and Phillips, 2019; Gonen and Goldberg, 2019; Hall Maudslay et al., 2019; Liang et al.,…”
Background
Research works printed but not linked: 3
- Clark, 2020, ELECTRA: Pre-training text encoders as discriminators rather than generators, The International Conference on Learning Representations (ICLR)
- Goldfarb-Tarrant, 2021, Intrinsic bias metrics do not correlate with application bias, 10.18653/v1/2021.acl-long.150 · DOI · Paper page
- Liu, 2019, RoBERTa: A robustly optimized BERT pretraining approach, CoRR
Cited by
Bias and Fairness in Large Language Models: A Survey
Gallegos, I. O., Rossi, R. A., Barrow, J., Tanjim, M. M., Kim, S., Dernoncourt, F., Yu, T., Zhang, R., & Ahmed, N. K. (2024) · Computational Linguistics · doi:10.1162/coli_a_00524 · Cited by 1,245
Biases in Large Language Models: Origins, Inventory, and Discussion
Navigli, R., Conia, S., & Ross, B. (2023) · Journal of Data and Information Quality · doi:10.1145/3597307 · Cited by 364
“Importantly, it has been found that the various sets of metrics used in hundreds of papers dealing with social bias can be uniied under three generalized fairness metrics: pairwise comparison, background comparison, and multi-group comparison metrics [41].”
BackgroundTowards trustworthy LLMs: a review on debiasing and dehallucinating in large language models
Lin, Z., Guan, S., Zhang, W., Zhang, H., Li, Y., & Zhang, H. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10896-y · Cited by 141
Human-LLM Collaborative Annotation Through Effective Verification of LLM Labels
Wang, X., Kim, H., Rahman, S., Mitra, K., & Miao, Z. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3641960 · Cited by 128
Policy advice and best practices on bias and fairness in AI
Álvarez, J. M., Colmenarejo, A. B., Elobaid, A., Fabbrizzi, S., Fahimi, M., Ferrara, A., Ghodsi, S., Mougan, C., Papageorgiou, I. T., Lobo, P. R., et al. (2024) · Ethics and Information Technology · doi:10.1007/s10676-024-09746-w · Cited by 160
“A few global events are targeted at multidisciplinary aspects of bias, fairness and other ethical issues in AI and algorithmic decision making.”
Language Model Behavior: A Comprehensive Survey
Chang, T. A., & Bergen, B. K. (2024) · Computational Linguistics · doi:10.1162/coli_a_00492 · Cited by 105
Interactive Model Cards: A Human-Centered Approach to Model Documentation
Crisan, A., Drouhard, M., Vig, J., & Rajani, N. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533108 · Cited by 101
“, metrics [16, 67]), and we urge caution to ensure that these approaches are understandable and actionable for non-experts.”
BackgroundIntegrating LLMs With ITS: Recent Advances, Potentials, Challenges, and Future Directions
Mahmud, D., Hajmohamed, H., Almentheri, S., Alqaydi, S., Aldhaheri, L., Khalil, R. A., & Saeed, N. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3528116 · Cited by 139
“Another effective method is the Background Comparison Metric (BCM), which evaluates model performance across different groups and highlights any disparities in accuracy [213].”
Robust Natural Language Processing: Recent Advances, Challenges, and Future Directions
Omar, M., Choi, S., Nyang, D., & Mohaisen, D. (2022) · IEEE Access · doi:10.1109/access.2022.3197769 · Cited by 85
“The fairness metric has been utilized by various studies [108], [109], [113], [114], [121] to evaluate the degree of NLP robustness and ensure that large models are fair.”
MethodWinoQueer: A community-in-the-loop benchmark for anti-LGBTQ+ bias in large language models
Felkner, V., Chang, H.-C. H., Jang, E., & May, J. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.507 · Cited by 74
You reap what you sow: On the Challenges of Bias Evaluation Under Multilingual Settings
Talat, Z., Névéol, A., Biderman, S., Clinciu, M., Dey, M., Longpre, S., Luccioni, S., Masoud, M., Mitchell, M., Radev, D., et al. (2022) · Proceedings of BigScience Episode #5 -- Workshop on Challenges & Perspectives in Creating Large Language Models · doi:10.18653/v1/2022.bigscience-1.3 · Cited by 59
“In response, previous work has explored ways in which such social biases can be mea-034 sured and counteracted (Nangia et al., 2020; Gehman et al., 2020; Czarnowska et al., 2021).”
BackgroundHate Speech Classifiers Learn Normative Social Stereotypes
Davani, A. M., Atari, M., Kennedy, B., & Dehghani, M. (2023) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00550 · Cited by 130
“As demonstrated by Czarnowska et al. (2021), various bias metrics can be considered as different parametrizations of a generalized metric.”
BackgroundNationality Bias in Text Generation
Narayanan Venkit, P., Gautam, S., Panchanadikar, R., Huang, T.-H., & Wilson, S. (2023) · Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2023.eacl-main.9 · Cited by 39
Challenges in Applying Explainability Methods to Improve the Fairness of NLP Models
Balkır, E., Kiritchenko, S., Nejadgholi, I., & Fraser, K. (2022) · Workshop on Trustworthy Natural Language Processing (TrustNLP) · doi:10.18653/v1/2022.trustnlp-1.8 · Cited by 67
Your fairness may vary: Pretrained language model fairness in toxic text classification
Baldini, I., Wei, D., Ramamurthy, K. N., Singh, M., & Yurochkin, M. (2022) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2022.findings-acl.176 · Cited by 78
“There are many definitions of group fairness and we refer to Verma and Rubin (2018) for a comprehensive overview and to Czarnowska et al. (2021) for a discussion of metrics in the context of measuring social biases in NLP. Statistical parity (SP) is one of the earlier definitions which requires the…”
BackgroundBozeman, J. F., Hollauer, C., Ramshankar, A. T., Nakkasunchi, S., Jambeck, J., Hicks, A., Bilec, M. M., McCauley, D., & Heidrich, O. (2024) · Journal of Industrial Ecology · doi:10.1111/jiec.13509 · Cited by 34
BERTScore is Unfair: On Social Bias in Language Model-Based Metrics for Text Generation
Sun, T., He, J., Qiu, X., & Huang, X. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.245 · Cited by 88
Toward Clinical-Grade Evaluation of Large Language Models
Moreno, A. C., & Bitterman, D. S. (2024) · International Journal of Radiation Oncology*Biology*Physics · doi:10.1016/j.ijrobp.2023.11.012 · Cited by 28
Data sharing and exchanging with incentive and optimization: a survey
Liu, L., & Han, M. (2024) · Discover Data · doi:10.1007/s44248-024-00006-2 · Cited by 24
“Addressing biases in natural language processing, specifically, those pertaining to gender, race, and age, is of particular importance [161–163].”
Dai, H., Mir, T. H., Chen, C.-T., Chen, C.-C., Yang, H., Lee, C.-H., Chou, Y.-Y., Teng, Y.-C., Gupta, S., Panchal, O., et al. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01921-7 · Cited by 28
“Finally, recent evidence 82 indicates that LLMs may exhibit unintended bias across demographic attributes.”
Gender Bias in Natural Language Processing and Computer Vision: A Comparative Survey
Bartl, M., Mandal, A., Leavy, S., & Little, S. (2024) · ACM Computing Surveys · doi:10.1145/3700438 · Cited by 48
“Individual fairness is achieved when changing the protected group, does not influence model output [35].”
Venkit, P. N., Srinath, M., & Wilson, S. (2023) · Workshop on Trustworthy Natural Language Processing (TrustNLP) · doi:10.18653/v1/2023.trustnlp-1.3 · Cited by 53
Persian offensive language detection
Kebriaei, E., Homayouni, A., Faraji, R., Razavi, A., Shakery, A., Faili, H., & Yaghoobzadeh, Y. (2023) · Machine Learning · doi:10.1007/s10994-023-06370-5 · Cited by 27
“For instance, Czarnowska et al. (2021) categorized the Fair-ness metrics into three generalized fairness metrics and compared different fairness metrics.”
A comparative study on the impact of model compression techniques on fairness in language models
Ramesh, K., Chavan, A., Pandit, S., & Sitaram, S. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.878 · Cited by 15
Das, D., Guha, S., Brubaker, J. R., & Semaan, B. (2024) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642669 · Cited by 49
“Such conversion of categorical outputs into a probability-based metric associated with the positive class for quantifying bias is common in NLP literature [51].”
(2024) · doi:10.1093/jamia/ocae001 · Cited by 14
Applying Large Language Models to Assess Quality of Care: Monitoring ADHD Medication Side Effects
Bannett, Y., Güntürkün, F., Pillai, M., Herrmann, J. E., Luo, I., Huffman, L. C., & Feldman, H. M. (2024) · PEDIATRICS · doi:10.1542/peds.2024-067223 · Cited by 14
On the independence of association bias and empirical fairness in language models
Cabello, L., Jørgensen, A. K., & Søgaard, A. (2023) · 2023 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3593013.3594004 · Cited by 14
SODAPOP: Open-Ended Discovery of Social Biases in Social Commonsense Reasoning Models
An, H., Li, Z., Zhao, J., & Rudinger, R. (2023) · Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2023.eacl-main.116 · Cited by 45
Scheuerman, M. K. (2024) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3630106.3658918 · Cited by 24
“…of commercial models (e.g., [2, 58, 64, 75]) and community benchmarks (e.g., [5, 6, 63]); developed taxonomies, frameworks, and toolkits (e.g., [11, 17, 26, 33, 41, 46]); presented sociotechnical analyses of the values present in AI and how they can lead to harms (e.g., [21, 45, 74]); and assessed…”
BackgroundComparing Biases and the Impact of Multilingual Training across Multiple Languages
Levy, S., John, N. A., Liu, L., Vyas, Y., Ma, J., Fujinuma, Y., Ballesteros, M., Castelli, V., & Roth, D. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.634 · Cited by 64
Self-citationThe tail wagging the dog: Dataset construction biases of social bias benchmarks
Selvam, N., Dev, S., Khashabi, D., Khot, T., & Chang, K.-W. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/2023.acl-short.118 · Cited by 11
This prompt is measuring <mask>: evaluating bias evaluation in language models
Goldfarb-Tarrant, S., Ungless, E. L., Balkır, E., & Blodgett, S. L. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.139 · Cited by 62
Easy Adaptation to Mitigate Gender Bias in Multilingual Text Classification
Huang, X. (2022) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/2022.naacl-main.52 · Cited by 15
Behind the Mask: Demographic bias in name detection for PII masking
Mansfield, C., Paullada, A., & Howell, K. (2022) · Workshop on Language Technology for Equality, Diversity, Inclusion (LT-EDI) · doi:10.18653/v1/2022.ltedi-1.10 · Cited by 20
Study on the Fairness of Speaker Verification Systems Across Accent and Gender Groups
Estévez, M., & Ferrer, L. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095150 · Cited by 16
“Most publications on fairness focus on tasks like face recognition [2, 3], natural language processing [4, 5], and automatic speech recognition [6, 7].”
BackgroundMeasuring Bias in a Ranked List Using Term-Based Representations
Abolghasemi, A., Azzopardi, L. A., Askari, A., de Rijke, M., & Verberne, S. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-56069-9_1 · Cited by 15
“While measuring gender bias with counterfactual data substitution is widely used for natural language processing tasks [9,14,30,42], we believe that our analysis falls short of thoroughly measuring the learned stereo-typical bias.”
BackgroundResponsible 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
“Fairness and Non-dis-crimination align with the principles of Justice and Fair-ness, ensuring that AA models do not perpetuate biases or discriminate against certain groups (Blodgett et al., 2020; Czarnowska et al., 2021; Halvani & Graner, 2021; Murauer & Specht, 2021a).”
Does ChatGPT show gender bias in behavior detection?
Wu, J., Song, Y., & Wu, D. C. (2024) · Humanities and Social Sciences Communications · doi:10.1057/s41599-024-04219-3 · Cited by 7
“Existing studies show the FP/FN-ED is an ideal choice to evaluate fairness in classi fi cation tasks (Czarnowska et al., 2021).”
The effects of gender bias in word embeddings on patient phenotyping in the mental health domain
Soğancıoğlu, G., Kaya, H., & Salah, A. A. (2023) · International Conference on Affective Computing and Intelligent Interaction and workshops · doi:10.1109/acii59096.2023.10388203 · Cited by 10
“The appropriate fairness measure for a model depends on its use case and the definition of fairness for such an application [36].”
BackgroundEpistemic consequences of unfair tools
Lassen, I. M. S., Kristensen‐McLachlan, R. D., Almasi, M., Enevoldsen, K., & Nielbo, K. L. (2024) · Digital Scholarship in the Humanities · doi:10.1093/llc/fqad091 · Cited by 9
Fonseca, G., da Cunha, W. L. M., & da Rocha, L. C. D. (2024) · Revista Eletrônica de Iniciação Científica em Computação · doi:10.5753/reic.2024.4643 · Cited by 5
Responsible AI Framework For Large Language Models (LLMs)
Kandhari, N., Tripathi, B., Kumar, S., Singh, K., & Singh, N. P. (2024) · International Conference on Advances in Computing and Communications (ICACC) · doi:10.1109/icacc63692.2024.10845690 · Cited by 4
Nejadgholi, I., Balkır, E., Fraser, K., & Kiritchenko, S. (2022) · BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP (BlackboxNLP) · doi:10.18653/v1/2022.blackboxnlp-1.18 · Cited by 13
T2IBias: Uncovering Societal Bias Encoded in the Latent Space of Text-to-Image Generative Models
Sufian, A., Distante, C., Leo, M., & Salam, H. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-16886-3_4 · Cited by 7
Wang, A., Bai, X., Barocas, S., & Blodgett, S. L. (2025) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3715275.3732046 · Cited by 10
Estratégias de Undersampling para Redução de Viés em Classificação de Texto Baseada em Transformers
Fonseca, G., Prenassi, G., da Cunha, W. L. M., Gonçalves, M. A., & da Rocha, L. C. D. (2024) · Brazilian Symposium on Multimedia and the Web · doi:10.5753/webmedia.2024.241229 · Cited by 8
Leveraging a Large Language Model to Assess Quality-of-Care: Monitoring ADHD Medication Side EffectsPreprint
Bannett, Y., Güntürkün, F., Pillai, M., Herrmann, J. E., Luo, I., Huffman, L. C., & Feldman, H. M. (2024) · medRxiv · doi:10.1101/2024.04.23.24306225 · Cited by 3
Bias Exposed: The BiaXposer Framework for NLP Fairness
Gaci, Y., Benatallah, B., Casati, F., & Benabdeslem, K. (2024) · Lecture notes in computer science · doi:10.1007/978-981-96-0805-8_22 · Cited by 4
Foundations of AI in Educational Assessment
Trajkovski, G., & Hayes, H. (2025) · Digital education and learning · doi:10.1007/978-3-031-88252-4_1 · Cited by 2
EquiSent: A Scalable Solution for Bias-Aware Gender-Fair Sentiment Modeling
Kundu, P., & Senthilselvi, A. (2025) · International Conference on Advances in Computation, Communication and Information Technology (ICAICCIT) · doi:10.1109/icaiccit68829.2025.11434462 · Cited by 1
Lost in Distillation: A Case Study in Toxicity Modeling
Chvasta, A., Lees, A., Sorensen, J. S., Vasserman, L., & Goyal, N. (2022) · Workshop on Online Abuse and Harms (WOAH) · doi:10.18653/v1/2022.woah-1.9 · Cited by 2
“Models with unintended biases has received considerable attention with multiple survey papers both generally (Pessach and Shmueli, 2022) and for natural language in particular (Kurita et al., 2019; Czarnowska et al., 2021).”
BackgroundAn, H., & Rudinger, R. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-short.34 · Cited by 21
Large language models in the labyrinth: Possibility spaces and moral constraints
Poulsen, V. M., & DeDeo, S. (2023) · Possibility Studies & Society · doi:10.1177/27538699231206210 · Cited by 3
“…and they also—for better or for worse—capture biases in human language, ranging from the cognitive biases studied in the psychological literature (Talboy & Fuller, 2023) to discriminatory attitudes encoded in the patterns of contemporary life (Bender et al., 2021; Czarnowska et al., 2021).”
BackgroundFantastic Biases (What are They) and Where to Find ThemPreprint
Barrière, V. (2024) · Qeios · doi:10.32388/ouxt01 · Cited by 1
Chen, H., Zhao, Z., & Ng, T. L. J. (2023) · Irish Conference on Artificial Intelligence and Cognitive Science (AICS) · doi:10.1109/aics60730.2023.10470824 · Cited by 1
“[9] discuss fairness metrics with a demographic focus, while [10] suggest training with demographically perturbed data to maintain fairness and performance.”
BackgroundImproving Commonsense Bias Classification by Mitigating the Influence of Demographic Terms
Lee, J., & Kim, J. (2024) · IEEE Access · doi:10.1109/access.2024.3477599
“Additionally, It has been shown that systematic differences in bias measurement can be revealed through the unification of extrinsic evaluation methods, attributed to parameter choices [12].”
BackgroundOmelchuk, S. A., Цапів, А. О., & Ставенко, О. В. (2025) · East European Journal of Psycholinguistics · doi:10.29038/ome
Bias Detection in Legal Texts Using Natural Language Processing
Gupta, R., Sapra, P., Tanwar, N., & Kavita (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-14189-7_2 · Cited by 1
Ensuring Equitable Guidance: A Context-Based Approach to Bias Detection in AI Advisors
Augello, A., Sabatucci, L., Neroni, P., Casoria, L., & Caggianese, G. (2025) · IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE) · doi:10.1109/metroxraine66377.2025.11340506
Okabe, S., Enomoto, T., Komachi, M., & Keyaki, A. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401573
“For instance, studies such as [14] and [4] have proposed methods for quantifying social bias within models, metrics for measuring biases in generated outputs, and techniques for mitigating such biases.”
Detecting and Mitigating Social Bias in Transformer Language Models
Dahatonde, R. T., Rashmi, R., Motghare, M., Shelke, N. M., Ashtankar, N., & Guda, S. (2025) · IEEE Global Conference on Wireless Computing and Networking (GCWCN) · doi:10.1109/gcwcn66157.2025.11448494
On Subjectivity, Bias and Fairness in Language Model Learning
Gaci, Y. (2023) · theses.fr (ABES) · doi:10.70675/dca82074za0c4z46d1z8a6bz96ce30bea3b6 · Cited by 3
Mitigating Implicit Bias in Chinese Toxic Speech Detection via Unbiased Contrastive Learning
Feng, X., Yao, Y., Gu, T., Chang, L., & Liu, X. (2026) · ACM Transactions on Asian and Low-Resource Language Information Processing · doi:10.1145/3816029 · Cited by 1
“It is important to note that the PCM is more reflective of bias values when model performance is low [12].”
DetBias: A Framework for Detecting Bias in NLP Classifier Models
Adil, M. F., Aswani, S., & Shetty, S. D. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-25617-1_32
Islam, M. S., Xiangdong, L., Ahmed, J., & Bing, X. (2026) · World Conference on Computing and Communication Technologies (WCCCT) · doi:10.1109/wccct69960.2026.11549734
“…study has examined gender or racial bias independently and highlight those biases are often compounded at their intersection, resulting in disproportionately lower performance for groups such as Black women [10, 11] strategies to ensure equitable sentiment analysis across languages and populations.”
Expert Consensus on LLM Safety in Clinical Decision Support
Gayathri, K. V., Kanshiya, S., Saranraj, P., Lokeshwari, B., Vilmasen, P., & Wins, J. A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-7862-6.ch011
Quantifying Social Biases in Language Model Classifiers is Domain-Dependent
Quiroga, T., Bravo-Márquez, F., & Barrière, V. (2026) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3834859
“We adopt two metrics described in Czarnowska et al. [13].”
Measuring and Mitigating Bias and Discrimination in AI
Ghalme, G. S. (2026) · Springer eBooks · doi:10.1007/978-3-032-22368-5_3
Dialect Bias in Arabic Toxicity Detection Under Controlled Distributional Shifts
Alamri, M. (2026) · Electronics · doi:10.3390/electronics15163612
Metadata
- Authors
- Czarnowska, P., Vyas, Y., & Shah, K.
- Affiliations
- University of Cambridge (GB)
- Amazon (United States) (US)
- Venue
- Transactions of the Association for Computational Linguistics
- Publisher
- MIT Press - Journals
- Published
- 2021
- Volume
- 9
- Pages
- 1249-1267
- ISSN
- 2307-387X (ISSN-L)
- OpenAlex
- W3174651299
- Cited by
- 72
- References
- 82
- Citations by year
- 2026: 9 · 2025: 10 · 2024: 22 · 2023: 19 · 2022: 12
- FWCI
- 16.59
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Transactions of the Association for Computational Linguistics (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
- Keywords
- fairness metrics, algorithmic bias, demographic bias, bias measurement, parameter selection
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.