Mitigating Age-Related Bias in Large Language Models: Strategies for Responsible Artificial Intelligence Development

References

• Adila D, Zhang S, Han B, Wang B (2024) Discovering bias in latent space: An unsupervised debiasing approach. Forty-First Internat. Conf. Machine Learn. (ICML 2024) (OpenReview.net).Google Scholar
• Agiza A, Mostagir M, Reda S (2024) PoliTune: Analyzing the impact of data selection and fine-tuning on economic and political biases in large language models. Preprint, submitted July 27, http://dx.doi.org/10.48550/ARXIV.2404.08699.Google Scholar
• Ba Y, Liu X, Chen X, Wang H, Xu Y, Li K, Zhang S (2024) Cautiously-optimistic knowledge sharing for cooperative multi-agent reinforcement learning. Wooldridge MJ, Dy JG, Natarajan S, eds. Thirty-Eighth AAAI Conf. Artificial Intelligence (AAAI Press, Palo Alto, CA), 17299–17307.Google Scholar
• Balvert M (2024) Iterative rule extension for logic analysis of data: An MILP-based heuristic to derive interpretable binary classifiers from large data sets. INFORMS J. Comput. 36(3):723–741.Link, Google Scholar
• Birru J, Chague F, De-Losso R, Giovannetti B (2024) Attention and biases: Evidence from tax-inattentive investors. Management Sci. 70(10):7101–7119.Link, Google Scholar
• Cai Y, Zhang C, Shen W, Zhang X, Ruan W, Huang L (2023) Reprem: Representation pre-training with masked model for reinforcement learning. Williams B, Chen Y, Neville J, eds. Thirty-Seventh AAAI Conf. Artificial Intelligence (AAAI Press, Palo Alto, CA), 6879–6887.Google Scholar
• Chen J, Liu L, Zhou F (2025) Do not wait: Preemptive rumor detection with cooperative LLMs and accessible social context. Inform. Processing Management 62(3):103995.Crossref, Google Scholar
• Chhikara G, Sharma A, Ghosh K, Chakraborty A (2024) Few-shot fairness: Unveiling LLM’s potential for fairness-aware classification. Preprint, submitted February 28, http://dx.doi.org/10.48550/ARXIV.2402.18502.Google Scholar
• Chu CH, Donato-Woodger S, Khan SS, Nyrup R, Leslie K, Lyn A, Shi T, Bianchi A, Rahimi SA, Grenier A (2023) Age-related bias and artificial intelligence: A scoping review. Humanities Soc. Sci. Comm. 10(1):510.Crossref, Google Scholar
• Dai S, Xu C, Xu S, Pang L, Dong Z, Xu J (2024) Bias and unfairness in information retrieval systems: New challenges in the LLM era. KDD 2024, 6437–6447.Google Scholar
• De Cremer D (2020) What does building a fair AI really entail. Harvard Bus. Rev. (September 3), https://hbr.org/2020/09/what-does-building-a-fair-ai-really-entail.Google Scholar
• Fan X, Hanasusanto GA (2024) A decision rule approach for two-stage data-driven distributionally robust optimization problems with random recourse. INFORMS J. Comput. 36(2):526–542.Link, Google Scholar
• Fernández-Ardèvol M, Grenier L (2024) Exploring data ageism: What good data can(‘t) tell us about the digital practices of older people? New Media Soc. 26(8):4611–4628.Crossref, Google Scholar
• Gallegos IO, Rossi RA, Barrow J, Tanjim MM, Kim S, Demoncourt F, Yu T, Zhang R, Ahmed NK (2024) Bias and fairness in large language models: A survey. Preprint, submitted July 12, http://dx.doi.org/10.48550/ARXIV.2309.00770.Google Scholar
• Ghanbarzadeh S, Huang Y, Palangi H, Moreno RC, Khanour H (2023) Gender-tuning: Empowering fine-tuning for debiasing pre-trained language models. Findings Association Computational Linguistics: ACL 2023 (Association for Computational Linguistic, Toronto), 5448–5458.Google Scholar
• Gu S, Knoll A, Jin M (2025) TeaMs-RL: Teaching LLMs to teach themselves better instructions via reinforcement learning. Preprint, submitted March 1, http://dx.doi.org/10.48550/ARXIV.2403.08694.Google Scholar
• Gupta S, Shriv V, Desh A, Kalyan A, Clark P, Sab A, Khot T (2024) Bias runs deep: Implicit reasoning biases in persona-assigned LLMs. Twelfth Internat. Conf. Learn. Representations (OpenReview.net).Google Scholar
• Gurevych I, Hovy EH, Slonim N, Stein B (2015) Debating technologies (Dagstuhl Seminar 15512). Dagstuhl Rep. 5(12):18–46.Google Scholar
• Haller P, Aynetdinov A, Akbik A (2024) OpinionGPT: Model. Explicit biases in instruction-tuned LLMs. Proc. 2024 Conf. North Amer. (Association for Computational Linguistics, Stroudsburg, PA), 78–86.Google Scholar
• Han Y (2024) Fairness evaluation within large language models through the lens of depression. Proc. 2023 4th Internat. Conf. Machine Learn. Comput. Appl. (Association for Computing Machinery, New York), 108–112.Google Scholar
• Harris C (2023) Mitigating age biases in resume screening AI models. Flairs 2023 (Clearwater Beach, FL).Google Scholar
• Harris CG (2024) Combining human-in-the-loop systems and AI fairness toolkits to reduce age bias in AI job hiring algorithms. BigComp 2024, 60–66.Google Scholar
• Hu J, Jiang Y, Weng P (2024) Revisiting data augmentation in deep reinforcement learning. Twelfth Internat. Conf. Learn. Representations (OpenReview.net).Google Scholar
• Hu B, Zhao C, Zhang P, Zhou Z, Yang Y, Xu Z, Liu B (2023) Enabling intelligent interactions between an agent and an LLM: A reinforcement learning approach. Preprint, submitted August 31, https://arxiv.org/abs/2306.03604v4.Google Scholar
• Jiang AQ, Sablayrolles A, Lacroix T, Sayed WE (2023) Mistral 7b. Preprint, submitted October 10, http://dx.doi.org/10.48550/ARXIV.2310.06825.Google Scholar
• Kamruzzaman M (2025) Investigating and mitigating undesirable biases in large language models. Walsh T, Shah J, Kolter Z, eds. AAAI-25, Sponsored Assoc. Advancement Artificial Intelligence (AAAI Press, Palo Alto, CA), 29273–29274.Google Scholar
• Kamruzzaman M, Shovon MMI, Kim GL (2024) Investigating subtler biases in LLMs: Ageism, beauty, institutional, and nationality bias in generative models. Ku LW, Martins A, Srikumar V, eds. Findings Association Computational Linguistics: ACL 2024 (Association for Computational Linguistic, Stroudsburg, PA), 8940–8965.Google Scholar
• Kelley S, Ovchinnikov A, Hardoon DR, Heinrich A (2022) Antidiscrimination laws, artificial intelligence, and gender bias: A case study in nonmortgage fintech lending. Manufacturing Service Oper. Management 24(6):3039–3059.Link, Google Scholar
• Kidder W, D’Cruz J, Varshney KR (2024) Empathy and the right to be an exception: What LLMs can and cannot do. Preprint, submitted January 25, http://dx.doi.org/10.48550/ARXIV.2401.14523.Google Scholar
• Kiehne N, Ljapunov A, Bätje M, Balke W (2024) Analyzing effects of learning downstream tasks on moral bias in LLMs. Calzolari N, Kan MY, Hoste V, Lenci A, Sakti S, Xue N, eds. Proc. 2024 Joint Internat. Conf. Comput. Linguistics, Language Resources Evaluation (ELRA and ICCL, Paris), 904–923.Google Scholar
• Kumar A, Yunusov S, Emami A (2024) Subtle biases need subtler measures: Dual metrics for evaluating representative and affinity bias in LLMs. Ku LW, Martins A, Srikumar V, eds. Proc. 62nd Annual Meeting Assoc. Comput. Linguistics, vol. 1 (Association for Computational Linguistic, Bangkok, Thailand), 375–392.Google Scholar
• Leslie D (2020) Tackling COVID-19 through responsible AI innovation: Five steps in the right direction. Harvard Data Sci Rev. (June 5), https://hdsr.mitpress.mit.edu/pub/as1p81um/release/3.Google Scholar
• Lin L, Wang L, Guo J, Wong K (2024) Investigating bias in LLM-based bias detection: Disparities between LLMs and human perception. Preprint, submitted December 10, http://dx.doi.org/10.48550/ARXIV.2403.14896.Google Scholar
• Liu S, Maturi T, Shen S, Mihalcea R (2024) The generation gap: Exploring age bias in large language models. Preprint, submitted October 15, http://dx.doi.org/10.48550/ARXIV.2404.08760.Google Scholar
• Liu Z, Qian S, Cao S, Shi T (2025) Mitigating age-related bias in large language models: Strategies for responsible artificial intelligence development. http://dx.doi.org/10.1287/ijoc.2024.0645.cd, https://github.com/INFORMSJoC/2024.0645.Google Scholar
• Liu Z, Huang D, Huang K, Li Z, Zhao J (2020) FinBERT: A pre-trained financial language representation model for financial text mining. Proc. Twenty-Ninth Internat. Joint Conf. Artificial Intelligence, IJCAI 2020 (ijcai.org), 4513–4519.Google Scholar
• Ma H, Zhang C, Bian Y, Liu L, Zhang Z, Zhao P, Zhang S (2023) Fairness-guided few-shot prompting for large language models. Preprint, submitted March 31, http://dx.doi.org/10.48550/ARXIV.2303.13217.Google Scholar
• Ma Y, Jiao L, Liu F, Li L, Ma W, Yang S, Liu X, Chen P (2025) Unveiling and mitigating generalized biases of DNNs through the intrinsic dimensions of perceptual manifolds. IEEE Trans. Pattern Anal. Machine Intelligence 47(3):2237–2244.Crossref, Google Scholar
• Maheshwari G, Bellet A, Denis P, Keller M (2023) Fair without leveling down: A new intersectional fairness definition. Bouamor H, Pino J, Kalika B, eds. Findings Assoc. Comput. Linguistics: EMNLP 2023 (Association for Computational Linguistics, Stroudsburg, PA), 9018–9032.Google Scholar
• Mak H (2022) Enabling smarter cities with operations management. Manufacturing Service Oper. Management 24(1):24–39.Link, Google Scholar
• Meade N, Gella S, Gupta P, Jin D, Reddy S, Liu Y (2023) Using in-context learning to improve dialogue safety. Bouamor H, Pino J, Kalika B, eds. Findings Assoc. Comput. Linguistics: EMNLP 2023 (Association for Computational Linguistics, Stroudsburg, PA), 11882–11910.Google Scholar
• Mehrabi N, Morstatter F, Saxena N, Lerman K, Galstyan A (2022) A survey on bias and fairness in machine learning. ACM Comput. Surveys 54(6):115:1–115:35.Crossref, Google Scholar
• Meta (2024) Llama 3. POPL ‘79 Proc. 6th ACM SIGACT-SIGPLAN Sympos. Principles Programming Languages, 226–236.Google Scholar
• Nangia N, Vania C, Bhalerao R, Bowman SR (2020) Crows-pairs: A challenge dataset for measuring social biases in masked LMs. Webber B, Cohn T, He Y, Liu Y, eds. Proc. 2020 Conf. Empirical Methods Natural Language Processing: EMNLP 2020 (Association for Computational Linguistics, Stroudsburg, PA), 1953–1967.Google Scholar
• Nguyen H, Eger S (2024) Is there really a citation age bias in NLP? Preprint, submitted January 7, http://dx.doi.org/10.48550/ARXIV.2401.03545.Google Scholar
• Oba D, Kaneko M, Bollegala D (2024) In-contextual gender bias suppression for large language models. Graham Y, Purver M, eds. Findings Association Computational Linguistics: EACL 2024 (Association for Computational Linguistic, Stroudsburg, PA), 1722–1742.Google Scholar
• Oketunji AF, Anas M, Saina D (2023) Large language model (LLM) bias index—LLMBI. Preprint, submitted December 29, http://dx.doi.org/10.48550/ARXIV.2312.14769.Google Scholar
• O’Leary DE (2025) Confirmation and specificity biases in large language models: An explorative study. IEEE Intelligent Systems 40(1):63–68.Crossref, Google Scholar
• OpenAI (2023) GPT-4 technical report. Preprint, submitted, http://dx.doi.org/10.48550/ARXIV.2303.08774.Google Scholar
• Parrish A, Chen A, Nangia N, Padmakumar V, Phang J, Thompson J, Htut PM, Bowman SR (2022) BBQ: A hand-built bias benchmark for question answering. Muresan S, Nakov P, Villavicencio A, eds. Findings Association Computational Linguistics: ACL 2024 (Association for Computational Linguistic, Stroudsburg, PA), 2086–2105.Google Scholar
• Peng Y, Xiao L, Hd B, Hong LJ, Lam H (2022) A new likelihood ratio method for training artificial neural networks. INFORMS J. Comput. 34(1):638–655.Link, Google Scholar
• Proebsting G, Poliak A (2025) Biases in large language model-elicited text: A case study in natural language inference. Rambow O, Wanner L, Apidianaki M, Al-Khalifa H, Di Eugenio B, Schockaert S, eds. Proc. 31st Internat. Conf. Comput. Linguistics: COLING 2025 (Association for Computational Linguistics, Stroudsburg, PA), 5836–5851.Google Scholar
• Rajabalizadeh A, Davarnia D (2024) Solving a class of cut-generating linear programs via machine learning. INFORMS J. Comput. 36(3):708–722.Link, Google Scholar
• Samorani M, Harris SL, Blount LG, Lu H, Santoro MA (2022) Overbooked and overlooked: Machine learning and racial bias in medical appointment scheduling. Manufacturing Service Oper. Management 24(6):2825–2842.Link, Google Scholar
• Shin J, Song H, Lee H, Jeong S, Park J (2025) Ask LLMs directly, “what shapes your bias?”: Measuring social bias in large language models. Ku LW, Martins A, Srikumar V, eds. Findings Association Computational Linguistics: ACL 2024 (Association for Computational Linguistic, Stroudsburg, PA), 16122–16143.Google Scholar
• Sun Y, Qi J, Zhu Z, Li K, Zhao L, Lv L (2025) Bias-guided margin loss for robust visual question answering. Inform. Processing Management 62(3):103988.Crossref, Google Scholar
• Tokpo EK, Calders T (2023) Model-based counterfactual generator for gender bias mitigation. Preprint, submitted November 6, http://dx.doi.org/10.48550/ARXIV.2311.03186.Google Scholar
• Touvron H, Martin L, Stone K, Albert P, Almahairi A, Babaei Y, Bashlykov N (2023) Llama 2: Open foundation and fine-tuned chat models. Preprint, submitted July 19, http://dx.doi.org/10.48550/ARXIV.2307.09288.Google Scholar
• Wang S, Delage E (2024) A column generation scheme for distributionally robust multi-item newsvendor problems. INFORMS J. Comput. 36(3):849–867.Link, Google Scholar
• Wu X, Nian J, Tao Z, Fang Y (2025) Evaluating social biases in LLM reasoning. Preprint, submitted February 21, http://dx.doi.org/10.48550/ARXIV.2502.15361.Google Scholar
• Xu W, Zhu G, Zhao X, Pan L, Li L, Wang W (2024) Pride and prejudice: LLM amplifies self-bias in self-refinement. Ku LW, Martins A, Srikumar V, eds. Findings Association Computational Linguistics: ACL 2024 (Association for Computational Linguistic, Stroudsburg, PA), 15474–15492.Google Scholar
• Xu Z, Chen W, Tang Y, Li X, Hu C, Chu Z, Ren K, Zheng Z, Lu Z (2025) Mitigating social bias in large language models: A multi-objective approach within a multi-agent framework. Walsh T, Shah J, Kolter Z, eds. AAAI-25, Sponsored Assoc. Advancement Artificial Intelligence (AAAI Press, Palo Alto, CA), 25579–25587.Google Scholar
• Yang C, Rustogi R, Wu T (2023) Beyond testers’ biases: Guiding model testing with knowledge bases using LLMs. Bouamor H, Pino J, Bali K, eds. Findings Assoc. Comput. Linguistics: EMNLP 2023 (Association for Computational Linguistics, Stroudsburg, PA), 13504–13519.Google Scholar
• Yang H, Wang Y, Xu X, Zhang H, Bian Y (2024a) Can we trust LLMs? Mitigate overconfidence bias in LLMs through knowledge transfer. Preprint, submitted May 27, http://dx.doi.org/10.48550/ARXIV.2405.16856.Google Scholar
• Yang A, Yang B, Hui B, Zheng B, Yu B, Zhou C, Li C (2024b) Qwen2 technical report. Preprint, submitted September 10, https://arxiv.org/abs/2407.10671.Google Scholar
• You Y, Huang J, Tong Q, Wang B (2025) Tackling biased complementary label learning with large margin. Inform. Sci. 687:121400.Crossref, Google Scholar
• Yu X, Shi R, Feng P, Tian Y, Li S, Liao S, Wu W (2024) Leveraging partial symmetry for multi-agent reinforcement learning. Wooldridge MJ, Dy JG, Natarajan S, eds. Thirty-Eighth AAAI Conf. Artificial Intelligence: AAAI 2024 (AAAI Press, Palo Alto, CA), 17583–17590.Google Scholar
• Zhang N, Xu H (2024) Fairness of ratemaking for catastrophe insurance: Lessons from machine learning. Inform. Systems Res. 35(2):469–488.Link, Google Scholar
• Zhao Y, Wang B, Wang Y (2025) Explicit vs. implicit: Investigating social bias in LLMs through self-reflection. Preprint, submitted March 7, http://dx.doi.org/10.48550/ARXIV.2501.02295.Google Scholar
• Zuccotto M, Castellini A, La Torre D, Mola L, Farinelli A (2024) Reinforcement learning applications in environmental sustainability: A review. Artificial Intelligence Rev. 57(4):88.Crossref, Google Scholar