The DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
BBQ: A hand-built bias benchmark for question answering
Open accessAbstract
It is well documented that NLP models learn social biases, but little work has been done on how these biases manifest in model outputs for applied tasks like question answering (QA).We introduce the Bias Benchmark for QA (BBQ), a dataset of question sets constructed by the authors that highlight attested social biases against people belonging to protected classes along nine social dimensions relevant for U.S. English-speaking contexts.Our task evaluates model responses at two levels: (i) given an under-informative context, we test how strongly responses refect social biases, and (ii) given an adequately informative context, we test whether the model's biases override a correct answer choice.We fnd that models often rely on stereotypes when the context is under-informative, meaning the model's outputs consistently reproduce harmful biases in this setting.Though models are more accurate when the context provides an informative answer, they still rely on stereotypes and average up to 3.4 percentage points higher accuracy when the correct answer aligns with a social bias than when it conficts, with this difference widening to over 5 points on examples targeting gender for most models tested.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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 source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“LLMs may exhibit social biases and toxicity [37, 53, 153] during the generation process, resulting in the production of biased outputs.”
From A Survey on Evaluation of Large Language Models (2024) · cites this paper as [153]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Evaluating the expanding capabilities of LLMs is a complex research area28,29, with group-specific bias benchmarks shown to be time-consuming to develop and utilize30–33, and the overall field lacking measurement validity and theoretical grounding30,34.”
From Generative language models exhibit social identity biases (2024) · cites this paper as [31]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Questions ask for a social group that meets either a negative or positive stereotype, with an ambiguous context that does not contain enough information to answer the question”
From Bias and Fairness in Large Language Models: A Survey (2024) · cites this paper as Parrish et al. 2022
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Bias Benchmark for QA (BBQ) (Parrish et al. 2022) is a questionanswering dataset of 58,492 examples to assess bias across nine social groups: age, disability status, gender, nationality, physical appearance, race/ethnicity, religion, and socioeconomic status.”
From Bias and Fairness in Large Language Models: A Survey (2024) · cites this paper as Parrish et al. 2022
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Provider documentation indicates that SES does appear in some evaluation suites. For example, BBQ (Bias Benchmark for QA), a standardized test measuring social biases in question-answering, includes SES dimensions [72]. Google's Gemini evaluations separately incorporate Dollar Street, a dataset of household photographs across income levels, for socioeconomic contexts [34, 84].”
From All Public Voices Are Equal, But Are Some More Equal Than Others to LLMs? (2026) · cites this paper as [72]
The cited source says what the claim attributes to it.
In the citing paper
“Question-answering models may rely on stereotypes to answer questions in ambiguous contexts (Dhamala et al. 2021; Parrish et al. 2022).”
From Bias and Fairness in Large Language Models: A Survey (2024) · cites this paper as Parrish et al. 2022
The cited source says what the claim attributes to it.
In the citing paper
“as well as a disambiguated context that specifies the correct answer.”
From Bias and Fairness in Large Language Models: A Survey (2024) · cites this paper as Parrish et al. 2022
The cited source says what the claim attributes to it.
In the citing paper
“The dataset is designed to measure when a model relies on stereotypes.”
From Bias and Fairness in Large Language Models: A Survey (2024) · cites this paper as Parrish et al. 2022
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
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019) · arXiv · doi:10.48550/arxiv.1907.11692 · Cited by 16,515
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
He, P., Gao, J., & Chen, W. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.09543 · Cited by 1,358
Language (Technology) is Power: A Critical Survey of “Bias” in NLP
Blodgett, S. L., Barocas, S., Daumé III, H., & Wallach, H. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.485 · Cited by 1,050 · cited 6×
“There are many, often conflicting, ways that bias is defined in NLP (Blodgett et al., 2020); we focus on stereotyping behavior and extend the definitions and treatment of bias in QA from Li et al. (2020), who have shown that the marginal probabilities a model associates with different answer…”
BackgroundInfluentialRACE: Large-scale ReAding Comprehension Dataset From Examinations
Lai, G., Xie, Q., Liu, H., Yang, Y., & Hovy, E. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1082 · Cited by 1,044
Gender bias in coreference resolution: Evaluation and debiasing methods
Zhao, J., Wang, T., Yatskar, M., Ordonez, V., & Chang, K.-W. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) · doi:10.18653/v1/n18-2003 · Cited by 684
Updating Occupational Prestige and Socioeconomic Scores: How the New Measures Measure up
Nakao, K., & Treas, J. (1994) · Sociological Methodology · doi:10.2307/270978 · Cited by 552
The Woman Worked as a Babysitter: On Biases in Language Generation
Sheng, E., Chang, K.-W., Natarajan, P., & Peng, N. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1339 · Cited by 537 · cited 2×
“LMs have been found to reproduce social biases in downstream tasks, such as textual entailment (Dev et al., 2020; Sharma et al., 2021), language generation (Sheng et al., 2019), and coreference resolution (Rudinger et al., 2018).”
BackgroundUNIFIEDQA: Crossing Format Boundaries with a Single QA System
Khashabi, D., Min, S., Khot, T., Sabharwal, A., Tafjord, O., Clark, P., & Hajishirzi, H. (2020) · Findings of the Association for Computational Linguistics: EMNLP 2020 · doi:10.18653/v1/2020.findings-emnlp.171 · Cited by 514
Social Bias Frames: Reasoning about Social and Power Implications of Language
Sap, M., Gabriel, S., Qin, L., Jurafsky, D., Smith, N. A., & Choi, Y. (2020) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2020.acl-main.486 · Cited by 469
Research works printed but not linked: 2
- Kate Crawford. 2017. The trouble with bias. http: //youtube.com/watch?v=fMym_BKWQzk. Talk given at NeurIPS December 2017.
- Daniel Khashabi, Tushar Khot, Ashish Sabharwal, Oyvind Tafjord, Peter Clark, and Hannaneh Ha- jishirzi. 2020. UnifedQA: Crossing format bound- aries with a single QA system. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Findings, pages 1896-1907.
Cited by
Sorted by citations
A Survey on Evaluation of Large Language Models
Chang, Y., Wang, X., Wang, J., Wu, Y., Yang, L., Zhu, K., Chen, H., Yi, X., Wang, C., Wang, Y., et al. (2024) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3641289 · Cited by 4,254
A Comprehensive Overview of Large Language Models
Naveed, H., Khan, A. U., Qiu, S., Saqib, M., Anwar, S., Usman, M., Akhtar, N., Barnes, N., & Mian, A. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3744746 · Cited by 1,257
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
Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
Bai, Y., Jones, A., Ndousse, K., Askell, A., Chen, A., DasSarma, N., Drain, D., Fort, S., Ganguli, D., Henighan, T., et al. (2022) · arXiv · doi:10.48550/arxiv.2204.05862 · Cited by 442
Self-citationMallen, A., Asai, A., Zhong, V., Das, R., Khashabi, D., & Hajishirzi, H. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.546 · Cited by 340
Metadata
- Authors
- Parrish, A., Chen, A., Nangia, N., Padmakumar, V., Phang, J., Thompson, J., Htut, P. M., & Bowman, S.
- Affiliations
- New York University (US)
- Venue
- Findings of the Association for Computational Linguistics: ACL 2022
- Published
- 2022
- Pages
- 2086-2105
- OpenAlex
- W3207316473
- Cited by
- 186
- References
- 34
- Citations by year
- 2026: 67 · 2025: 62 · 2024: 29 · 2023: 23 · 2022: 5
- FWCI
- 29.56
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Findings of the Association for Computational Linguistics: ACL (Findings of ACL) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- National Science Foundation (2046556, 1922658)
- Samsung
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
- Keywords
- social bias, question answering, bias benchmark, stereotypes, protected classes, gender bias, model bias evaluation, large language models
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
