From Lexicons to Large Language Models: A Holistic Evaluation of Psychometric Text Analysis in Social Science Research

References

• Abbasi A, Zhou Y, Deng S, Zhang P (2018) Text analytics to support sense-making in social media: A language-action perspective. MIS Quart. 42(2):427–464.Crossref, Google Scholar
• Abbasi A, Jeffrey P, Gautam P, Olivia S, Suprateek S (2024) Pathways for design research on artificial intelligence. Inform. Systems Res. 35(2):441–459.Link, Google Scholar
• Abbasi A, Dobolyi D, Lalor JP, Netemeyer RG, Smith K, Yang Y (2021) Constructing a psychometric testbed for fair natural language processing. Proc. 2021 Conf. Empirical Methods Natural Language Processing (Association for Computational Linguistics, Stroudsburg, PA), 3748–3758.Google Scholar
• Adamopoulos P, Ghose A, Todri V (2018) The impact of user personality traits on word of mouth: Text-mining social media platforms. Inform. Systems Res. 29(3):612–640.Link, Google Scholar
• Ahmad F, Abbasi A, Li J, Dobolyi DG, Netemeyer RG, Clifford GD, Chen H (2020) A deep learning architecture for psychometric natural language processing. ACM Trans. Inform. Systems 38(1):1–29.Crossref, Google Scholar
• Althammer S, Zuccon G, Hofstätter S, Verberne S, Hanbury A (2023) Annotating data for fine-tuning a neural ranker? Current active learning strategies are not better than random selection. Proc. Annual Internat. ACM SIGIR-AP ‘23 (Association for Computing Machinery, New York), 139–149.Google Scholar
• Anglin AH, Kincaid PA, Short JC, Allen DG (2022) Role theory perspectives: Past, present, and future applications of role theories in management research. J. Management 48(6):1469–1502.Crossref, Google Scholar
• Antoniak M, Mimno D (2021) Bad seeds: Evaluating lexical methods for bias measurement. Proc. 59th Annual Meeting Assoc. Comput. Linguistics and 11th Internat. Joint Conf. Natl. Language Processing, vol. 1: Long Papers (Association for Computational Linguistics, Stroudsburg, PA), 1889–1904.Google Scholar
• Araci D (2019) FinBERT: Financial sentiment analysis with pre-trained language models. Preprint, submitted August 27, https://arxiv.org/abs/1908.10063.Google Scholar
• Arseniev-Koehler A, Foster JG (2022) Machine learning as a model for cultural learning: Teaching an algorithm what it means to be fat. Sociol. Methods Res. 51(4):1484–1539.Crossref, Google Scholar
• Baccianella S, Esuli A, Sebastiani F (2010) SentiWordNet 3.0: An enhanced lexical resource for sentiment analysis and opinion mining. Proc. Seventh Internat. Conf. Language Resources Evaluation LREC’10 (European Language Resources Association, Paris), 2200–2204.Google Scholar
• Barbey AK, Colom R, Paul EJ, Grafman J (2014) Architecture of fluid intelligence and working memory revealed by lesion mapping. Brain Structure Function 219(2):485–494.Crossref, Google Scholar
• Batson D (2016) Empathy and altruism. Brown KW, Leary MR, eds. The Oxford Handbook of Hypo-Egoic Phenomena (Oxford University Press, Oxford, UK), 161–174.Google Scholar
• Beltagy I, Lo K, Cohan A (2019) SciBERT: A pretrained language model for scientific text. Preprint, submitted September 10, https://arxiv.org/abs/1903.10676.Google Scholar
• Bhardwaj R, Majumder N, Poria S (2021) Investigating gender bias in BERT. Cognitive Comput. 13(4):1008–1018.Crossref, Google Scholar
• Biddle BJ, Thomas EJ (1966) Role Theory: Concepts and Research (John Wiley & Sons, Hoboken, NJ).Google Scholar
• Black RC, Treul SA, Johnson TR, Goldman J (2011) Emotions, oral arguments, and Supreme Court decision making. J. Politics 73(2):572–581.Crossref, Google Scholar
• Bras RL, Swayamdipta S, Bhagavatula C, Zellers R, Peters ME, Sabharwal A, Choi Y (2020) Adversarial filters of dataset biases. Preprint, submitted July 11, https://arxiv.org/abs/2002.04108.Google Scholar
• Brown T, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, Neelakantan A, et al. (2020) Language models are few-shot learners. Adv. Neural Inform. Processing Systems 33:1877–1901.Google Scholar
• Cella D, Lai JS, Nowinski CJ, Victorson D, Peterman A, Miller D, Bethoux F, et al. (2012) Neuro-QOL: Brief measures of health-related quality of life for clinical research in neurology. Neurology 78(23):1860–1867.Crossref, Google Scholar
• Chen T, Guestrin C (2016) XGBoost: A scalable tree boosting system. Proc. 22nd ACM SIGKDD Internat. Conf. Knowledge Discovery Data Mining (Association for Computing Machinery, New York), 785–794.Google Scholar
• Chen J, Liu Z, Huang X, Wu C, Liu Q, Jiang G, Pu Y, et al. (2023) When large language models meet personalization: Perspectives of challenges and opportunities. Preprint, submitted July 31, https://arxiv.org/abs/2307.16376.Google Scholar
• Chen J, Wang X, Xu R, Yuan S, Zhang Y, Shi W, Xie J, et al. (2024) From persona to personalization: A survey on role-playing language agents. Preprint, submitted October 9, https://arxiv.org/abs/2404.18231.Google Scholar
• Chia YK, Chen G, Tuan LA, Poria S, Bing L (2023) Contrastive chain-of-thought prompting. Preprint, submitted November 15, https://arxiv.org/abs/2311.09277.Google Scholar
• Chowdhery A, Narang S, Devlin J, Bosma M, Mishra G, Roberts A, Barham P, et al. (2022) PaLM: Scaling language modeling with pathways. Preprint, submitted October 5, https://arxiv.org/abs/2204.02311.Google Scholar
• Clore GL, Huntsinger JR (2007) How emotions inform judgment and regulate thought. Trends Cognitive Sci. 11(9):393–399.Crossref, Google Scholar
• Cortes C, Vapnik V (1995) Support-vector networks. Machine Learn. 20(3):273–297.Crossref, Google Scholar
• Crawford JR, Henry JD (2004) The Positive and Negative Affect Schedule (PANAS): Construct validity, measurement properties and normative data in a large non-clinical sample. British J. Clin. Psych. 43(3):245–265.Crossref, Google Scholar
• Decety J, Jackson PL (2004) The functional architecture of human empathy. Behav. Cognitive Neurosci. Rev. 3(2):71–100.Crossref, Google Scholar
• de Lima FF, Osório FdL (2021) Empathy: Assessment instruments and psychometric quality—A systematic literature review with a meta-analysis of the past ten years. Front Psychol. 12:781346.Crossref, Google Scholar
• De Neys W (2018) Dual Process Theory 2.0 (Routledge/Taylor & Francis Group, New York).Google Scholar
• Devlin J, Chang MW, Lee K, Toutanova K (2019) BERT: Pre-training of deep bidirectional transformers for language understanding. Proc. 2019 Conf. North Amer. Chapter Assoc. Comput. Linguistics: Human Language Technologies, vol. 1: Long and Short Papers (Association for Computational Linguistics, Stroudsburg, PA), 4171–4186.Google Scholar
• Dror R, Baumer G, Shlomov S, Reichart R (2018) The Hitchhiker’s guide to testing statistical significance in natural language processing. Gurevych I, Miyao Y, eds. Proc. 56th Annual Meeting Assoc. Comput. Linguistics, vol. 1: Long Papers (Association for Computational Linguistics, Stroudsburg, PA), 1383–1392.Google Scholar
• Eloundou T, Beutel A, Robinson DG, Gu-Lemberg K, Brakman AL, Mishkin P, Shah M, Heidecke J, Weng L, Kalai AT (2024) First-person fairness in chatbots. Preprint, submitted October 16, https://arxiv.org/abs/2410.19803.Google Scholar
• Evans J (2003) In two minds: Dual-process accounts of reasoning. Trends Cognitive Sci. 7(10):454–459.Crossref, Google Scholar
• Evans J, Stanovich KE (2013) Dual-process theories of higher cognition: Advancing the debate. Perspect. Psych. Sci. 8(3):223–241.Crossref, Google Scholar
• Fan T, Wang H, Hodel T (2023) Multimodal knowledge graph construction of Chinese traditional operas and sentiment and genre recognition. J. Cultural Heritage 62:32–44.Crossref, Google Scholar
• Fedus W, Zoph B, Shazeer N (2022) Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity. Preprint, submitted June 16, https://arxiv.org/abs/2101.03961.Google Scholar
• Fiori M (2009) A new look at emotional intelligence: A dual-process framework. Personality Soc. Psych. Rev. 13(1):21–44.Crossref, Google Scholar
• Fiori M, Ortony A (2021) Initial evidence for the hypersensitivity hypothesis: Emotional intelligence as a magnifier of emotional experience. J. Intelligence 9(2):24.Crossref, Google Scholar
• Friedler SA, Scheidegger C, Venkatasubramanian S, Choudhary S, Hamilton EP, Roth D (2018) A comparative study of fairness-enhancing interventions in machine learning. Preprint, submitted February 12, https://arxiv.org/abs/1802.04422.Google Scholar
• Fröhling L, Bernardelle P, Demartini G (2024) SubData: A Python library to collect and combine datasets for evaluating LLM alignment on downstream tasks. Preprint, submitted December 21, https://arxiv.org/abs/2412.16783.Google Scholar
• Gamache D, McNamara G (2019) Responding to bad press: How CEO temporal focus influences the sensitivity to negative media coverage of acquisitions. Acad. Management J. 62(3):918–943.Crossref, Google Scholar
• Gershon RC, Lai JS, Bode R, Choi S, Moy C, Bleck T, Miller D, Peterman A, Cella D (2012) Neuro-QOL: Quality of life item banks for adults with neurological disorders: Item development and calibrations based upon clinical and general population testing. Quality Life Res. 21(3):475–486.Crossref, Google Scholar
• Gneiting T, Raftery AE (2007) Strictly proper scoring rules, prediction, and estimation. J. Amer. Statist. Assoc. 102(477):359–378.Crossref, Google Scholar
• Grattafiori A, Dubey A, Jauhri A, Pandey A, Kadian A, Al-Dahle A, Letman A, et al. (2024) The Llama 3 herd of models. Preprint, submitted November 23, https://arxiv.org/abs/2407.21783.Google Scholar
• Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput. 9(8):1735–1780.Crossref, Google Scholar
• Hu EJ, Shen Y, Wallis P, Allen-Zhu Z, Li Y, Wang S, Wang L, Chen W (2021) LoRA: Low-rank adaptation of large language models. Preprint, submitted October 16, https://arxiv.org/abs/2106.09685.Google Scholar
• Huang AH, Wang H, Yang Y (2023) FinBERT: A large language model for extracting information from financial text. Contemporary Accounting Res. 40(2):806–841.Crossref, Google Scholar
• Hyde S, Bachura E, Bundy J, Gretz R, Sanders G (2024) The tangled webs we weave: Examining the effects of CEO deception on analyst recommendations. Strategic Management J. 45(1):66–112.Crossref, Google Scholar
• Inselberg A (2009) Parallel coordinates. Liu L, Tamer Özsu, eds. Encyclopedia of Database Systems (Springer US, New York), 2018–2024.Crossref, Google Scholar
• Ji J, Qiu T, Chen B, Zhang B, Lou H, Wang K, Duan Y, et al. (2024) AI alignment: A comprehensive survey. Preprint, submitted May 1, https://arxiv.org/abs/2310.19852.Google Scholar
• Jiang AQ, Sablayrolles A, Mensch A, Bamford C, Chaplot DS, de las Casas D, Bressand F, et al. (2023) Mistral 7B. Preprint, submitted October 10, https://arxiv.org/abs/2310.06825.Google Scholar
• Joseph DL, Newman DA (2010) Emotional intelligence: An integrative meta-analysis and cascading model. J. Appl. Psych. 95(1):54–78.Crossref, Google Scholar
• Kahneman D (2011) Thinking, Fast and Slow (Farrar, Straus and Giroux, New York).Google Scholar
• Kahneman D, Frederick S (2002) Representativeness revisited: Attribute substitution in intuitive judgment. Gilovich T, Griffin D, Kahneman D, eds. Heuristics and Biases: The Psychology of Intuitive Judgment (Cambridge University Press, New York), 49–81.Crossref, Google Scholar
• Khemlani SS, Barbey AK, Johnson-Laird PN (2014) Causal reasoning with mental models. Front. Human Neurosci. 8:849.Crossref, Google Scholar
• Kitchens B, Claggett J, Abbasi A (2024) Timely, granular, and actionable: Designing a social listening platform for public health 3.0. MIS Quart. 48(3):899–930.Crossref, Google Scholar
• Kong A, Zhao S, Chen H, Li Q, Qin Y, Sun R, Zhou X, Wang E, Dong X (2024) Better zero-shot reasoning with role-play prompting. Preprint, submitted March 14, https://arxiv.org/abs/2308.07702.Google Scholar
• König A, Graf-Vlachy L, Bundy J, Little LM (2020) A blessing and a curse: How CEOs’ trait empathy affects their management of organizational crises. Acad. Management Rev. 45(1):130–153.Crossref, Google Scholar
• Lee K, Ram S (2024) Explainable deep learning for false information identification: An argumentation theory approach. Inform. Systems Res. 35(2):890–907.Link, Google Scholar
• Lee D, Cheng Z, Mao C, Manzoor E (2025) Guided Diverse Concept Miner (GDCM): Uncovering relevant constructs for managerial insights from text. Inform. Systems Res. 36(1):370–393.Link, Google Scholar
• Lester B, Al-Rfou R, Constant N (2021) The power of scale for parameter-efficient prompt tuning. Preprint, submitted September 2, https://arxiv.org/abs/2104.08691.Google Scholar
• Liu Y, Pant G, Sheng OR (2020) Predicting labor market competition: Leveraging interfirm network and employee skills. Inform. Systems Res. 31(4):1443–1466.Link, Google Scholar
• Liu X, Wang GA, Fan W, Zhang Z (2020) Finding useful solutions in online knowledge communities: A theory-driven design and multilevel analysis. Inform. Systems Res. 31(3):731–752.Link, Google Scholar
• Liu S, Liu S, Liu Z, Peng X, Yang Z (2022) Automated detection of emotional and cognitive engagement in MOOC discussions to predict learning achievement. Comput. Ed. 181:104461.Crossref, Google Scholar
• Liu S, Zhang Z, Yan R, Wu W, Yang C, Lu J (2024) Measuring spiritual values and bias of large language models. Preprint, submitted October 15, https://arxiv.org/abs/2410.11647.Google Scholar
• Loughran T, McDonald B (2011) When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. J. Finance 66(1):35–65.Crossref, Google Scholar
• Madaan A, Tandon N, Gupta P, Hallinan S, Gao L, Wiegreffe S, Alon U, et al. (2023) Self-refine: Iterative refinement with self-feedback. Preprint, submitted May 25, https://arxiv.org/abs/2303.17651.Google Scholar
• Mehrabi N, Morstatter F, Saxena N, Lerman K, Galstyan A (2022) A survey on bias and fairness in machine learning. Preprint, submitted January 25, https://arxiv.org/abs/1908.09635.Google Scholar
• Miller GA (1995) WordNet: A lexical database for English. Comm. ACM 38(11):39–41.Crossref, Google Scholar
• Miranda S, Nicholas B, Stefan S, Hani S, Andrew BJ (2022) Computationally intensive theory construction: A primer for authors and reviewers. MIS Quart. 46(2):3–18.Google Scholar
• Mishra S, Arunkumar A, Sachdeva B, Bryan C, Baral C (2020) DQI: Measuring data quality in NLP. Preprint, submitted May 2, https://arxiv.org/abs/2005.00816.Google Scholar
• Mousavi R, Gu B (2024) Resilience messaging: The effect of governors’ social media communications on community compliance during a public health crisis. Inform. Systems Res. 35(2):505–527.Link, Google Scholar
• Neal A, Ballard T, Vancouver JB (2017) Dynamic self-regulation and multiple-goal pursuit. Annual Rev. Organ. Psych. Organ. Behav. 4(1):401–423.Crossref, Google Scholar
• Panksepp J (2003) At the interface of the affective, behavioral, and cognitive neurosciences: Decoding the emotional feelings of the brain. Brain Cognition 52(1):4–14.Crossref, Google Scholar
• Pennebaker JW, Francis ME (1996) Cognitive, emotional, and language processes in disclosure. Cognition Emotion 10(6):601–626.Crossref, Google Scholar
• Peters H, Matz S (2023) Large language models can infer psychological dispositions of social media users. Preprint, submitted September 13, https://arxiv.org/abs/2309.08631.Google Scholar
• Podsakoff PM, MacKenzie SB, Lee JY, Podsakoff NP (2003) Common method biases in behavioral research: A critical review of the literature and recommended remedies. J. Appl. Psych. 88(5):879–903.Crossref, Google Scholar
• Qiao M, Huang KW (2021) Correcting misclassification bias in regression models with variables generated via data mining. Inform. Systems Res. 32(2):462–480.Link, Google Scholar
• Rathje S, Mirea DM, Sucholutsky I, Marjieh R, Robertson C, Bavel JJV (2023) GPT is an effective tool for multilingual psychological text analysis. Preprint, submitted May 19, https://osf.io/sekf5.Google Scholar
• Rozado D (2020) Wide range screening of algorithmic bias in word embedding models using large sentiment lexicons reveals underreported bias types. PLoS One 15(4):e0231189.Crossref, Google Scholar
• Rumelhart DE, Smolensky P, McClelland JL, Hinton GE (1986) Sequential thought processes in PDP models. Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol. 2: Psychological and Biological Models (MIT Press, Cambridge, MA), 7–57.Google Scholar
• Rust RT, Rand W, Huang MH, Stephen AT, Brooks G, Chabuk T (2021) Real-time brand reputation tracking using social media. J. Marketing 85(4):21–43.Crossref, Google Scholar
• Salovey P, Mayer JD (1990) Emotional intelligence. Imagination Cognition Personality 9(3):185–211.Crossref, Google Scholar
• Salton G, Wong A, Yang CS (1975) A vector space model for automatic indexing. Comm. ACM 18(11):613–620.Crossref, Google Scholar
• Sedoc J, Buechel S, Nachmany Y, Buffone A, Ungar L (2020) Learning word ratings for empathy and distress from document-level user responses. Proc. Twelfth Language Resources Evaluation Conf. (European Language Resources Association, Paris), 1664–1673.Google Scholar
• Sergent K, Stajkovic AD (2020) Women’s leadership is associated with fewer deaths during the COVID-19 crisis: Quantitative and qualitative analyses of United States governors. J. Appl. Psych. 105(8):771–783.Crossref, Google Scholar
• Shazeer N, Mirhoseini A, Maziarz K, Davis A, Le Q, Hinton G, Dean J (2017) Outrageously large neural networks: The sparsely-gated mixture-of-experts layer. Preprint, submitted January 23, https://arxiv.org/abs/1701.06538.Google Scholar
• Shen C, Xie G, Zhang X, Xu J (2024) On the decision-making abilities in role-playing using large language models. Preprint, submitted February 29, https://arxiv.org/abs/2402.18807.Google Scholar
• Singer T, Lamm C (2009) The social neuroscience of empathy. Ann. NY Acad. Sci. 1156:81–96.Crossref, Google Scholar
• Slovic P, Finucane ML, Peters E, MacGregor DG (2007) The affect heuristic. Eur. J. Oper. Res. 177(3):1333–1352.Crossref, Google Scholar
• Stanovich KE, West RF (1998) Individual differences in rational thought. J. Experiment. Psych. General 127(2):161–188.Crossref, Google Scholar
• Stone PJ, Dunphy DC, Smith MS (1966) The General Inquirer: A Computer Approach to Content Analysis (MIT Press, Oxford, UK).Google Scholar
• Strapparava C, Mihalcea R (2007) SemEval-2007 Task 14: Affective text. Agirre E, Màrquez L, Wicentowski R, eds. Proc. Fourth Internat. Workshop Semantic Evaluations SemEval-2007 (Association for Computational Linguistics, Stroudsburg, PA), 70–74.Google Scholar
• Suslow T, Hoepfel D, Günther V, Kersting A, Bodenschatz CM (2022) Positive attentional bias mediates the relationship between trait emotional intelligence and trait affect. Sci. Rep. 12(1):20733.Crossref, Google Scholar
• Tan Z, Beigi A, Wang S, Guo R, Bhattacharjee A, Jiang B, Karami M, Li J, Cheng L, Liu H (2024) Large language models for data annotation: A survey. Preprint, submitted December 2, https://arxiv.org/abs/2402.13446.Google Scholar
• Tausczik YR, Pennebaker JW (2010) The psychological meaning of words: LIWC and computerized text analysis methods. J. Language Soc. Psych. 29(1):24–54.Crossref, Google Scholar
• Toma CL, Hancock JT (2012) What lies beneath: The linguistic traces of deception in online dating profiles. J. Comm. 62(1):78–97.Crossref, Google Scholar
• Truninger M, Fernández-I-Marín X, Batista-Foguet JM, Boyatzis RE, Serlavós R (2018) The power of EI competencies over intelligence and individual performance: A task-dependent model. Front. Psych. 9:1532.Crossref, Google Scholar
• Tseng YM, Huang YC, Hsiao TY, Chen WL, Huang CW, Meng Y, Chen YN (2024) Two tales of persona in LLMs: A survey of role-playing and personalization. Preprint, submitted October 5, https://arxiv.org/abs/2406.01171.Google Scholar
• Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Ł, Polosukhin I (2017) Attention is all you need. 31st Conf. Neural Inform. Processing Systems (NIPS 2017) (Curran Associates, Inc., Red Hook, NY).Google Scholar
• Wei J, Wang X, Schuurmans D, Bosma M, Ichter B, Xia F, Chi E, Le Q, Zhou D (2023) Chain-of-thought prompting elicits reasoning in large language models. Preprint, submitted January 10, https://arxiv.org/abs/2201.11903.Google Scholar
• Wei J, Tay Y, Bommasani R, Raffel C, Zoph B, Borgeaud S, Yogatama D, et al. (2022) Emergent abilities of large language models. Preprint, submitted October 26, https://arxiv.org/abs/2206.07682.Google Scholar
• Willemsen J, Vanheule S, Verhaeghe P (2011) Psychopathy and lifetime experiences of depression. Criminal Behaviour Mental Health 21(4):279–294.Crossref, Google Scholar
• Wong CS, Law KS (2002) Wong and Law Emotional Intelligence Scale (WLEIS) [Database record]. APA PsycTests. https://doi.org/10.1037/t07398-000.Google Scholar
• Wu N, Gong M, Shou L, Liang S, Jiang D (2023) Large language models are diverse role-players for summarization evaluation. Preprint, submitted September 19, https://arxiv.org/abs/2303.15078.Google Scholar
• Xu Y, Armony M, Ghose A (2021) The interplay between online reviews and physician demand: An empirical investigation. Management Sci. 67(12):7344–7361.Link, Google Scholar
• Xu L, Xie H, Qin SZJ, Tao X, Wang FL (2023) Parameter-efficient fine-tuning methods for pretrained language models: A critical review and assessment. Preprint, submitted December 19, https://arxiv.org/abs/2312.12148.Google Scholar
• Yang K, Lau RYK, Abbasi A (2023) Getting personal: A deep learning artifact for text-based measurement of personality. Inform. Systems Res. 34(1):194–222.Link, Google Scholar
• Yang Y, Zhang K, Fan Y (2023) SDTM: A supervised Bayesian deep topic model for text analytics. Inform. Systems Res. 34(1):137–156.Link, Google Scholar
• Yang M, Adomavicius G, Burtch G, Ren Y (2018) Mind the gap: Accounting for measurement error and misclassification in variables generated via data mining. Inform. Systems Res. 29(1):4–24.Link, Google Scholar
• Yao S, Yu D, Zhao J, Shafran I, Griffiths TL, Cao Y, Narasimhan K (2023) Tree of thoughts: Deliberate problem solving with large language models. Preprint, submitted December 3, https://arxiv.org/abs/2305.10601.Google Scholar
• Zhang D, Zhou L, Tao J, Zhu T, Gao G (2024) KETCH: A knowledge-enhanced transformer-based approach to suicidal ideation detection from social media content. Inform. Systems Res. 36(1):572–599.Link, Google Scholar
• Zheng M, Pei J, Logeswaran L, Lee M, Jurgens D (2024) When “a helpful assistant” is not really helpful: Personas in system prompts do not improve performances of large language models. Preprint, submitted October 9, https://arxiv.org/abs/2311.10054.Google Scholar