{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,3]],"date-time":"2026-11-03T17:33:04Z","timestamp":1793727184328,"version":"4.6.1"},"reference-count":35,"publisher":"SAGE Publications","issue":"3","license":[{"start":{"date-parts":[[2026,7,11]],"date-time":"2026-07-11T00:00:00Z","timestamp":1783728000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"funder":[{"DOI":"10.13039\/501100004147","name":"Tsinghua University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100004147","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["WORK: A Journal of Prevention, Assessment &amp; Rehabilitation"],"published-print":{"date-parts":[[2026,11]]},"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Large language models (LLMs) harbor systematic biases that are particularly consequential in workplace and HR contexts, where their outputs increasingly influence hiring, job design, and organizational decisions. Yet existing bias-evaluation approaches remain methodologically fragmented, limiting practitioners\u2019 ability to assess deployment risks.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study introduces BiasLab, a multilingual dual-framing framework to quantify and compare directional output-level bias in LLMs, demonstrated across six workplace and HR-relevant topics.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>BiasLab combines mirrored affirmative and reverse prompt pairs, randomized wrapper perturbations, fixed-choice response constraints, and polarity-aligned scoring. Ten LLMs were evaluated across six topics, gender in leadership, employment gap candidates, age in hiring, remote versus office work, four-day versus five-day work weeks, and AI-assisted versus human-only hiring, spanning 12 languages and 30 iterations per framing direction, yielding 43,200 responses.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>All ten models showed consistent directional preferences across every topic: favoring female managers, gap candidates as equally capable, older workers, remote work, the four-day week, and AI-assisted hiring. A recurring asymmetric pattern emerged in which models rejected disfavored claims more strongly than they endorsed their opposites, a distinction invisible to single-frame designs.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>BiasLab provides a standardized, reproducible instrument for measuring directional preferences across models. Whether a preference constitutes bias in a fairness sense is topic-dependent: for protected attributes such as gender and age it maps onto equal-employment standards, whereas elsewhere it is better described as systematic preference. These findings have direct implications for HR decision-making, and the framework lets organizations compare and vet models before adopting them for hiring.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1177\/10519815261469150","type":"journal-article","created":{"date-parts":[[2026,7,11]],"date-time":"2026-07-11T12:12:12Z","timestamp":1783771932000},"page":"707-721","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":0,"title":["BiasLab: A multilingual dual-framing framework for LLM bias measurement, applied to workplace and HR contexts"],"prefix":"10.1177","volume":"85","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-7248-9361","authenticated-orcid":false,"given":"William","family":"Guey","sequence":"first","affiliation":[{"name":"Department of Industrial Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8208-3342","authenticated-orcid":false,"given":"Wei","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Industrial Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5713-8612","authenticated-orcid":false,"given":"Pei-Luen Patrick","family":"Rau","sequence":"additional","affiliation":[{"name":"Department of Industrial Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-1352-8108","authenticated-orcid":false,"given":"Pierrick","family":"Bougault","sequence":"additional","affiliation":[{"name":"Department of Industrial Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-0047-3647","authenticated-orcid":false,"given":"Vitor D","family":"de Moura","sequence":"additional","affiliation":[{"name":"School of Social Sciences, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-1106-0822","authenticated-orcid":false,"given":"Bertan","family":"Ucar","sequence":"additional","affiliation":[{"name":"Department of Industrial Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6838-714X","authenticated-orcid":false,"given":"Jos\u00e9 O","family":"Gomes","sequence":"additional","affiliation":[{"name":"Department of Industrial Engineering, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2026,7,11]]},"reference":[{"key":"e_1_3_3_2_2","doi-asserted-by":"publisher","DOI":"10.1145\/3605943"},{"key":"e_1_3_3_3_2","first-page":"374","volume-title":"Large language models in finance: a survey4th ACM international conference on AI in finance","author":"Li Y","unstructured":"Li Y, Wang S, Ding H, et al. Large language models in finance: a survey. In: 4th ACM international conference on AI in finance, 2023, pp.374\u2013382: ACM."},{"key":"e_1_3_3_4_2","doi-asserted-by":"publisher","DOI":"10.3390\/informatics11030057"},{"key":"e_1_3_3_5_2","first-page":"4776","volume-title":"Large language models in education: vision and opportunities2023 IEEE international conference on big data (BigData)","author":"Gan W","unstructured":"Gan W, Qi Z, Wu J, et al. Large language models in education: vision and opportunities. In: 2023 IEEE international conference on big data (BigData), pp.4776\u20134785: IEEE."},{"key":"e_1_3_3_6_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41562-025-02242-1"},{"key":"e_1_3_3_7_2","doi-asserted-by":"publisher","unstructured":"Guo Y Guo M Su J et al. Bias in large language models: origin evaluation and mitigation. arXiv [Preprint] 16 November 2024. DOI: 10.48550\/arXiv.2411.10915.","DOI":"10.48550\/arXiv.2411.10915"},{"key":"e_1_3_3_8_2","first-page":"49","volume-title":"Understanding intrinsic socioeconomic biases in large language modelsProceedings of the 2024 AAAI\/ACM conference on AI, ethics, and society","author":"Arzaghi M","unstructured":"Arzaghi M, Carichon F, Farnadi G. Understanding intrinsic socioeconomic biases in large language models. In: Proceedings of the 2024 AAAI\/ACM conference on AI, ethics, and society, 2025, pp.49\u201360: AAAI Press."},{"key":"e_1_3_3_9_2","doi-asserted-by":"publisher","DOI":"10.1038\/s43588-024-00741-1"},{"key":"e_1_3_3_10_2","doi-asserted-by":"publisher","DOI":"10.1186\/s13643-025-03000-0"},{"key":"e_1_3_3_11_2","first-page":"102082","article-title":"Bias in AI-driven HRM systems: investigating discrimination risks embedded in AI recruitment tools and HR analytics","volume":"12","author":"Sony MMAAM","year":"2025","unstructured":"Sony MMAAM, Amin MB, Ashraf A, et al. Bias in AI-driven HRM systems: investigating discrimination risks embedded in AI recruitment tools and HR analytics. Soc Sci Humanit Open 2025; 12: 102082.","journal-title":"Soc Sci Humanit Open"},{"key":"e_1_3_3_12_2","doi-asserted-by":"publisher","DOI":"10.3390\/app142411750"},{"key":"e_1_3_3_13_2","first-page":"2740","volume-title":"Gender bias in masked language models for multiple languagesProceedings of the 2022 conference of the North American chapter of the association for computational linguistics: human language technologies","author":"Kaneko M","unstructured":"Kaneko M, Imankulova A, Bollegala D, et al. Gender bias in masked language models for multiple languages. In: Proceedings of the 2022 conference of the North American chapter of the association for computational linguistics: human language technologies, 2022, pp.2740\u20132750: Association for Computational Linguistics."},{"key":"e_1_3_3_14_2","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0306621"},{"key":"e_1_3_3_15_2","doi-asserted-by":"publisher","unstructured":"Guey W Bougault P de Moura VD et al. Mapping geopolitical bias in 11 large language models: a bilingual dual-framing analysis of U.S.-China Tensions. arXiv [Preprint] 31 March 2025. DOI: 10.48550\/arXiv.2503.23688.","DOI":"10.48550\/arXiv.2503.23688"},{"key":"e_1_3_3_16_2","doi-asserted-by":"publisher","DOI":"10.1515\/jtc-2023-0019"},{"key":"e_1_3_3_17_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01746-4"},{"key":"e_1_3_3_18_2","doi-asserted-by":"publisher","DOI":"10.1162\/coli_a_00558"},{"key":"e_1_3_3_19_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2416228122"},{"key":"e_1_3_3_20_2","first-page":"56","article-title":"Enhancing fairness in LLM evaluations: unveiling and mitigating biases in standard-answer-based evaluations","volume":"4","author":"Jiao T","year":"2024","unstructured":"Jiao T, Zhang J, Xu K, et al. Enhancing fairness in LLM evaluations: unveiling and mitigating biases in standard-answer-based evaluations. Proc AAAI Symp Ser 2024; 4: 56\u201359.","journal-title":"Proc AAAI Symp Ser"},{"key":"e_1_3_3_21_2","doi-asserted-by":"publisher","unstructured":"Elbouanani A Dufraisse E Tuo A et al. CEA-LIST at CheckThat! 2025: evaluating LLMs as detectors of bias and opinion in text. arXiv [Preprint] 10 July 2025. DOI: 10.48550\/arXiv.2507.07539.","DOI":"10.48550\/arXiv.2507.07539"},{"key":"e_1_3_3_22_2","doi-asserted-by":"publisher","unstructured":"Shafiei M Saffari H Moosavi NS. More or less wrong: a benchmark for directional bias in LLM comparative reasoning. arXiv [Preprint] 4 June 2025. DOI: 10.48550\/arXiv.2506.03923.","DOI":"10.48550\/arXiv.2506.03923"},{"key":"e_1_3_3_23_2","doi-asserted-by":"publisher","DOI":"10.1145\/3597307"},{"key":"e_1_3_3_24_2","doi-asserted-by":"publisher","DOI":"10.1162\/coli_a_00524"},{"key":"e_1_3_3_25_2","doi-asserted-by":"publisher","unstructured":"Karinshak E Hu A Kong K et al. LLM-GLOBE: a benchmark evaluating the cultural values embedded in LLM output. arXiv [Preprint] 9 November 2024. DOI: 10.48550\/arXiv.2411.06032.","DOI":"10.48550\/arXiv.2411.06032"},{"key":"e_1_3_3_26_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2305016120"},{"key":"e_1_3_3_27_2","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.3628"},{"key":"e_1_3_3_28_2","first-page":"3227","volume-title":"Jobfair: a framework for benchmarking gender hiring bias in large language modelsFindings of the association for computational linguistics: EMNLP","author":"Wang Z","unstructured":"Wang Z, Wu Z, Guan X, et al. Jobfair: a framework for benchmarking gender hiring bias in large language models. In: Findings of the association for computational linguistics: EMNLP, 2024, pp.3227\u20133246."},{"key":"e_1_3_3_29_2","doi-asserted-by":"publisher","DOI":"10.1093\/pnasnexus\/pgaf089"},{"key":"e_1_3_3_30_2","doi-asserted-by":"publisher","DOI":"10.1017\/S004727941300055X"},{"key":"e_1_3_3_31_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-025-09581-z"},{"key":"e_1_3_3_32_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41562-025-02259-6"},{"key":"e_1_3_3_33_2","doi-asserted-by":"publisher","unstructured":"Wilson K Caliskan A. Gender race and intersectional bias in resume screening via language model retrieval. [Preprint] 20 August 2024. DOI: 10.48550\/arXiv.2407.20371.","DOI":"10.48550\/arXiv.2407.20371"},{"key":"e_1_3_3_34_2","doi-asserted-by":"publisher","DOI":"10.1609\/aies.v8i3.36749"},{"key":"e_1_3_3_35_2","unstructured":"Regulation (EU) 2024\/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence and amending Regulations (EC) No 300\/2008 (EU) No 167\/2013 (EU) No 168\/2013 (EU) 2018\/858 (EU) 2018\/1139 and (EU) 2019\/2144 and Directives 2014\/90\/EU (EU) 2016\/797 and (EU) 2020\/1828 (Artificial Intelligence Act) (Text with EEA relevance) http:\/\/data.europa.eu\/eli\/reg\/2024\/1689\/oj (2024 accessed 31 May 2026)."},{"key":"e_1_3_3_36_2","unstructured":"New York City Council. Local law 144 of 2021: automated employment decision tools https:\/\/www.lawandtheworkplace.com\/wp-content\/uploads\/sites\/29\/2023\/03\/Local-Law-144.pdf (2021)."}],"container-title":["WORK: A Journal of Prevention, Assessment &amp; Rehabilitation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/10519815261469150","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/10519815261469150","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/10519815261469150","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,11,3]],"date-time":"2026-11-03T16:46:21Z","timestamp":1793724381000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/10519815261469150"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,11]]},"references-count":35,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,11]]}},"alternative-id":["10.1177\/10519815261469150"],"URL":"https:\/\/doi.org\/10.1177\/10519815261469150","relation":{},"ISSN":["1051-9815","1875-9270"],"issn-type":[{"value":"1051-9815","type":"print"},{"value":"1875-9270","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,11]]}}}