mlsanityData to 7 Oct 2026

In-Depth Look at Word Filling Societal Bias Measures

arXiv 2302.12640 ↗First posted 24 Feb 2023cs.CLcited by 8

Not on the map. Why?

Abstract (arXiv)

Many measures of societal bias in language models have been proposed in recent years. A popular approach is to use a set of word filling prompts to evaluate the behavior of the language models. In this work, we analyze the validity of two such measures -- StereoSet and CrowS-Pairs. We show that these measures produce unexpected and illogical results when appropriate control group samples are constructed. Based on this, we believe that they are problematic and using them in the future should be reconsidered. We propose a way forward with an improved testing protocol. Finally, we also introduce a new gender bias dataset for Slovak.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 26 of its 43 references that are in the map. First posted on arXiv 24 Feb 2023.

Builds on

26 of 43 references.

References · 26

  1. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  2. 2020Language (Technology) is Power: A Critical Survey of "Bias" in NLPcited by 951
  3. 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 911
  4. 2016Semantics derived automatically from language corpora contain human-like biasescited by 894
  5. 2020CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Modelscited by 703
  6. 2018Gender Bias in Coreference Resolutioncited by 438
  7. 2019On Measuring Social Biases in Sentence Encoderscited by 397
  8. 2019Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Settingcited by 346
  9. 2019Measuring Bias in Contextualized Word Representationscited by 275
  10. 2019Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Flycited by 240
  11. 2020Measuring and Reducing Gendered Correlations in Pre-trained Modelscited by 222
  12. 2020Intrinsic Bias Metrics Do Not Correlate with Application Biascited by 148
14 more
  1. 2019Predictive Biases in Natural Language Processing Models: A Conceptual Framework and Overviewcited by 137
  2. 2021Harms of Gender Exclusivity and Challenges in Non-Binary Representation in Language Technologiescited by 128
  3. 2021An Empirical Survey of the Effectiveness of Debiasing Techniques for Pre-trained Language Modelscited by 96
  4. 2021A Survey on Gender Bias in Natural Language Processingcited by 89
  5. 2021Mitigating Language-Dependent Ethnic Bias in BERTcited by 87
  6. 2022On the Intrinsic and Extrinsic Fairness Evaluation Metrics for Contextualized Language Representationscited by 75
  7. 2022Theories of "Gender" in NLP Bias Researchcited by 62
  8. 2022Gender Bias in Masked Language Models for Multiple Languagescited by 60
  9. 2021Frequency Effects on Syntactic Rule Learning in Transformerscited by 59
  10. 2022Choose Your Lenses: Flaws in Gender Bias Evaluationcited by 29
  11. 2022Challenges in Measuring Bias via Open-Ended Language Generationcited by 23
  12. 2021Measuring Fairness with Biased Rulers: A Survey on Quantifying Biases in Pretrained Language Modelscited by 20
  13. 2022The AI Index 2022 Annual Reportcited by 18
  14. 2021SlovakBERT: Slovak Masked Language Modelcited by 11

Built on by

Cited by year

Citing papers per year: 2023: 4; 2024: 2; 2025: 2.

8 papers.

  1. 2023Social Bias Probing: Fairness Benchmarking for Language Modelscited by 29
  2. 2023Sociodemographic Bias in Language Models: A Survey and Forward Pathcited by 27
  3. 2023Women Are Beautiful, Men Are Leaders: Gender Stereotypes in Machine Translation and Language Modelingcited by 12
  4. 2024How far can bias go? Tracing bias from pretraining data to alignmentcited by 4
  5. 2023All Should Be Equal in the Eyes of Language Models: Counterfactually Aware Fair Text Generationcited by 3
  6. 2024Local Contrastive Editing of Gender Stereotypescited by 3
  7. 2025EsBBQ and CaBBQ: The Spanish and Catalan Bias Benchmarks for Question Answeringcited by 1
  8. 2025Social Bias in Multilingual Language Models: A Surveycited by 1