mlsanityData to 14 Oct 2026

Using Natural Sentences for Understanding Biases in Language Models

arXiv 2205.06303 ↗First posted 12 May 2022cs.CL cs.LGcited by 19

Not on the map. Why?

Abstract (arXiv)

Evaluation of biases in language models is often limited to synthetically generated datasets. This dependence traces back to the need for a prompt-style dataset to trigger specific behaviors of language models. In this paper, we address this gap by creating a prompt dataset with respect to occupations collected from real-world natural sentences present in Wikipedia. We aim to understand the differences between using template-based prompts and natural sentence prompts when studying gender-occupation biases in language models. We find bias evaluations are very sensitive to the design choices of template prompts, and we propose using natural sentence prompts for systematic evaluations to step away from design choices that could introduce bias in the observations.

Key facts. Cited by 19 papers in the map (data through 14 Oct 2026). Builds on 6 of its 10 references that are in the map. First posted on arXiv 12 May 2022.

Builds on

6 of 10 references.

References · 6

  1. 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 659
  2. 2019The Woman Worked as a Babysitter: On Biases in Language Generationcited by 541
  3. 2021BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generationcited by 412
  4. 2018Analysis Methods in Neural Language Processing: A Surveycited by 277
  5. 2018Gender Bias in Neural Natural Language Processingcited by 211
  6. 2020Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Biascited by 191

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 2; 2023: 9; 2024: 3; 2025: 2; 2026: 3.

19 papers.

Sort:

By area:

  1. 2023Language Model Behavior: A Comprehensive Surveycited by 92
  2. 2023Bias Against 93 Stigmatized Groups in Masked Language Models and Downstream Sentiment Classification Taskscited by 51
  3. 2023MISGENDERED: Limits of Large Language Models in Understanding Pronounscited by 43
  4. 2022Geographic and Geopolitical Biases of Language Modelscited by 29
  5. 2023Sociodemographic Bias in Language Models: A Survey and Forward Pathcited by 28
  6. 2023Automated Ableism: An Exploration of Explicit Disability Biases in Sentiment and Toxicity Analysis Modelscited by 20
  7. 2024Robust Pronoun Fidelity with English LLMs: Are they Reasoning, Repeating, or Just Biased?cited by 17
  8. 2023Fairness in Language Models Beyond English: Gaps and Challengescited by 16
  9. 2023CALM : A Multi-task Benchmark for Comprehensive Assessment of Language Model Biascited by 14
  10. 2025What Makes a Good Natural Language Prompt?cited by 7
  11. 2023Revealing the Unwritten: Visual Investigation of Beam Search Trees to Address Language Model Prompting Challengescited by 6
  12. 2024generAItor: Tree-in-the-Loop Text Generation for Language Model Explainability and Adaptationcited by 5
7 more
  1. 2025Mind the Language Gap: Automated and Augmented Evaluation of Bias in LLMs for High- and Low-Resource Languagescited by 4
  2. 2026Auditing Preferences for Brands and Cultures in LLMscited by 4
  3. 2026Regional Bias in Large Language Modelscited by 2
  4. 2022Gender Biases Unexpectedly Fluctuate in the Pre-training Stage of Masked Language Modelscited by 1
  5. 2024Template-Based Probes Are Imperfect Lenses for Counterfactual Bias Evaluation in LLMscited by 1
  6. 2023Multilingual Text Representationcited by 0
  7. 2026Exposing the Unsaid: Visualizing Hidden LLM Bias through Stochastic Path Aggregationcited by 0