Elamur Research⌕
☰
home / papers / 2025 / Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English

Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English

Paper · 2025 · Bryce Anderson, Riley Galpin and Tom S. Juzek

Summary

The paper asks whether the well-documented surge of "LLM-flavoured" words such as delve or intricate only shows people pasting AI-generated text, or whether it signals a change in how humans themselves produce language. Written and scripted texts may now be partly machine-made, so the authors use spontaneous speech instead. They built a 22.1-million-word corpus from conversational science and technology podcasts. The audio was transcribed with Whisper and tagged with spaCy. They then compared the frequency per million words of 20 AI-associated lemma–POS items before and after ChatGPT's 2022 release. A set of synonym baseline words served as a control. Most target words became more frequent, and several increases were statistically significant. The baseline words moved up and down in roughly equal numbers, with no clear direction. The authors read this as early evidence that human word choice may be drifting toward LLM style. They leave open whether this is ordinary language change or AI-driven influence. They link the question to alignment, since preferences introduced during Learning from Human Feedback could "seep" into human usage.

Key points

  • The corpus contains 22.1M words of unscripted podcast speech. Transcription used openai-whisper (about 92 hours of runtime), tagging used spaCy 3.8.2, and statistics used scipy. Code and data are at github.com/fsu-nlp/ai-traces-ssl.
  • The 34 target words come from Galpin, Anderson, and Juzek (2025), and 20 of them appear in the corpus. Of the 117 baseline synonyms, 87 appear.
  • 14 of the 20 target words increased after 2022. Significant rises were found for surpass (+140.8%), boast (+140.1%), strategically (+87.9%), align (+36.6%) and significant (+17.4%).
  • Two target words fell significantly: potential ADJ (−17.6%) and realm (−24.6%). Delve rose by about 47% (2.93 → 4.30 per million), but this change was not significant.
  • The authors also note a "delve-pushback" in PubMed abstracts. Frequency fell from about 22 to about 15 per million between 2024 and Q1 2025, possibly because the word now carries an AI stigma.

Interesting ideas

  • I1. Pick a data source where machine authorship is unlikely, here unscripted speech, because written text now suffers from "human-authorship indeterminacy" and is a weaker proxy for human behaviour.
  • I2. Pair every target word with matched synonym baselines, so that a real effect has to beat the background drift of comparable vocabulary.
  • I3. Count items as lemma plus POS (potential NOUN vs ADJ, showcase VERB vs NOUN) and normalise to occurrences per million, which keeps homographs apart and makes corpora of different sizes comparable.
  • I4. Treat overused "AI-style" vocabulary as a measurable stylistic bias, scored as target-word rate against a baseline.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

  • I1 (learn): Silver corpora and teacher-tier data could carry a provenance flag marking pre-LLM, spoken, or possibly AI-written origin, so that evaluation sets stay free of LLM-style contamination.
  • I2 (learn): Any claimed improvement from repair rules or self-play could be checked against a matched control set of similar constructions, to show the gain is specific rather than a general shift.
  • I3 (tokens): Frequency and lexicon statistics in crate en could be keyed by lemma+UPOS and stored as per-million rates, which would make tagger drift across domains easy to compare.
  • I4 (back): Text generated from the meaning graph could be audited for LLM-marked words such as delve, intricate or pivotal, preferring neutral synonyms so the output reads as human and stays faithful to the source style.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Florida State UniversityTallahassee · United States8 works · 363 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksHuman-Computer Interaction · 141 works

🧭 Topics 2Research subjects and languages the work is about.English · 2,263 worksHistorical linguistics & language change · 249 works

Cite

Cite as Bryce Anderson, Riley Galpin and Tom S. Juzek. 2025. Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English. arXiv:2508.00238.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2508.00238 · arXiv 2508.00238

@misc{anderson2025model,
  title = {{Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English}},
  author = {Bryce Anderson, Riley Galpin and Tom S. Juzek},
  year = {2025},
  doi = {10.48550/arxiv.2508.00238},
  eprint = {2508.00238},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2508.00238}
}

References

103 references, 15 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Aitchison, J. 2005. Language Change. In Cobley, P., ed., The Routledge Companion to Semiotics and Linguistics, chapter 7. London: Routledge. find
  2. Akintande, O. J. 2023. Algorithmic Bias: When stigmatization becomes a perception: The stigmatized become endangered. In Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society, 966–971. find
  3. Al-Sharqi, L.; and Abbasi, I. S. 2020. The Influence of Technology on English Language and Literature. English Language Teaching, 13(7): 1. ↗
  4. Alemohammad, S.; Casco-Rodriguez, J.; Luzi, L.; Humayun, A. I.; Babaei, H.; LeJeune, D.; Siahkoohi, A.; and Baraniuk, R. G. 2023. Self-consuming generative models go mad. arXiv preprint arXiv:2307.01850. ↗
  5. Ali, J.; Kleindessner, M.; Wenzel, F.; Budhathoki, K.; Cevher, V.; and Russell, C. 2023. Evaluating the fairness of discriminative foundation models in computer vision. In Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society, 809–833. find
  6. Amato, R.; Lacasa, L.; Dı́az-Guilera, A.; and Baronchelli, A. 2018. The dynamics of norm change in the cultural evolution of language. Proceedings of the National Academy of Sciences, 115(33): 8260–8265. ↗
  7. Bao, T.; Zhao, Y.; Mao, J.; and Zhang, C. 2025. Examining linguistic shifts in academic writing before and after the launch of ChatGPT: a study on preprint papers. Scientometrics, 1–31. ↗
  8. Bender, E. M.; Gebru, T.; McMillan-Major, A.; and Shmitchell, S. 2021. On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM conference on fairness, accountability, and transparency, 610–623. find
  9. Biber, D. 1991. Variation across speech and writing. Cambridge university press. find
  10. Bland, M. 2015. An introduction to medical statistics. Oxford university press. find
  11. Blodgett, S. L.; Barocas, S.; Daumé III, H.; and Wallach, H. 2020. Language (technology) is power: A critical survey of” bias” in nlp. arXiv preprint arXiv:2005.14050. ↗
  12. Bochkarev, V.; Solovyev, V.; and Wichmann, S. 2014. Universals versus historical contingencies in lexical evolution. Journal of the Royal Society Interface, 11(101): 20140841. ↗
  13. Bolukbasi, T.; Chang, K.-W.; Zou, J. Y.; Saligrama, V.; and Kalai, A. T. 2016. Man is to computer programmer as woman is to homemaker? debiasing word embeddings. Advances in neural information processing systems, 29. ↗
  14. Briesch, M.; Sobania, D.; and Rothlauf, F. 2023. Large language models suffer from their own output: An analysis of the self-consuming training loop. arXiv preprint arXiv:2311.16822. ↗
  15. Bybee, J. L. 2006. From usage to grammar: The mind’s response to repetition. Language, 82(4): 711–733. ↗
  16. Chen, Y.; Zhao, W.; Breitbarth, A.; Stoeckel, M.; Mehler, A.; and Eger, S. 2024. Syntactic language change in english and german: Metrics, parsers, and convergences. arXiv preprint arXiv:2402.11549. ↗
  17. Christiano, P. F.; Leike, J.; Brown, T.; Martic, M.; Legg, S.; and Amodei, D. 2017. Deep reinforcement learning from human preferences. Advances in neural information processing systems, 30. ↗
  18. Clark, H. H. 1996. Using Language. Cambridge University Press. find
  19. Clark, T. H.; Poliak, M.; Regev, T.; Haskins, A. J.; Gibson, E.; and Robertson, C. 2025. The relationship between surprisal, prosody, and backchannels in conversation reflects intelligibility-oriented pressures. PsyArXiv preprints. ↗
  20. Comas-Forgas, R.; Koulouris, A.; and Kouis, D. 2025. ‘AI-navigating’or ‘AI-sinking’? An analysis of verbs in research articles titles suspicious of containing AI-generated/assisted content. Learned Publishing, 38(1): e1647. find
  21. Crystal, D. 2009. Txtng: The Gr8 Db8. find
  22. Djalolovna, M. S. 2025. The Influence of Technology on Spoken Language: Texting, Voice Assistants, and Speech Patterns. Web of Technology: Multidimensional Research Journal, 3(1): 34–38. find
  23. Eisenstein, E. 1980. The Printing Press as an Agent of Change: Communications and Cultural Transformations in Early-Modern Europe. Cambridge University Press. ↗
  24. Fedorenko, E.; Piantadosi, S. T.; and Gibson, E. A. F. 2024. Language is primarily a tool for communication rather than thought. Nature, 630(8017): 575–586. ↗
  25. Fillmore, C. J. 2006. Frame Semantics. In Geeraerts, D.; Dirven, R.; and Taylor, J. R., eds., Cognitive Linguistics: Basic Readings, volume 34 of Cognitive Linguistics Research, 373–400. Berlin: Mouton de Gruyter. find
  26. Floridi, L. 2025. Distant Writing: Literary Production in the Age of Artificial Intelligence. Minds and Machines, 35(3): 1–26. ↗
  27. Floridi, L.; Cowls, J.; Beltrametti, M.; Chatila, R.; Chazerand, P.; Dignum, V.; Luetge, C.; Madelin, R.; Pagallo, U.; Rossi, F.; et al. 2018. AI4People—an ethical framework for a good AI society: opportunities, risks, principles, and recommendations. Minds and machines, 28: 689–707. ↗
  28. Gabriel, I. 2020. Artificial intelligence, values, and alignment. Minds and machines, 30(3): 411–437. ↗
  29. Galpin, R.; Anderson, B.; and Juzek, T. S. 2025. Lexical underrepresentation matters as much as lexical overrepresentation. arXiv preprint arXiv:2501.20200. ↗
  30. Gee, J. P. 2014. An introduction to discourse analysis: Theory and method. routledge. ↗
  31. Geng, M.; Chen, C.; Wu, Y.; Chen, D.; Wan, Y.; and Zhou, P. 2024. The Impact of Large Language Models in Academia: from Writing to Speaking. arXiv preprint arXiv:2409.13686. ↗
  32. Ghosh, S.; and Caliskan, A. 2023. Chatgpt perpetuates gender bias in machine translation and ignores non-gendered pronouns: Findings across bengali and five other low-resource languages. In Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society, 901–912. ↗
  33. Goffman, E. 1981. Forms of talk. University of Pennsylvania Press. find
  34. Grabowicz, P.; Perello, N.; and Takatsu, K. 2023. Learning from Discriminatory Training Data. In Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society, 752–763. find
  35. Gray, A. 2024. ChatGPT” contamination”: estimating the prevalence of LLMs in the scholarly literature. arXiv preprint arXiv:2403.16887. ↗
  36. Hagen, L.; Huo, J.; and Nguyen, A. 2025. Elon Musk’s AI Chatbot, Grok, Started Calling Itself ’MechaHitler’. NPR. find
  37. Hartvigsen, T.; Gabriel, S.; Palangi, H.; Sap, M.; Ray, D.; and Kamar, E. 2022. Toxigen: A large-scale machine-generated dataset for adversarial and implicit hate speech detection. arXiv preprint arXiv:2203.09509. ↗
  38. Haslwanter, T. 2016. An introduction to statistics with python. With applications in the life sciences. Switzerland: Springer International Publishing. ↗
  39. Hataya, R.; Bao, H.; and Arai, H. 2023. Will large-scale generative models corrupt future datasets? In Proceedings of the IEEE/CVF International Conference on Computer Vision, 20555–20565. find
  40. Helen, J. W. 2010. The Effect of Internet Devices on Children’s Language Development. Contemporary Issues in Communication Science Disorders, 37: 141–148. find
  41. Hendrycks, D.; Burns, C.; Basart, S.; Critch, A.; Li, J.; Song, D.; and Steinhardt, J. 2020. Aligning ai with shared human values. arXiv preprint arXiv:2008.02275. ↗
  42. Huang, B.; Chen, C.; and Shu, K. 2025. Authorship attribution in the era of llms: Problems, methodologies, and challenges. ACM SIGKDD Explorations Newsletter, 26(2): 21– 43. ↗
  43. Jaiswal, S. 2024. Uncovering Gender Biases in Human-AI Platforms. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, volume 7(2), 21–22. find
  44. Jin, H.; Sung, S.; Park, S.; Baik, S.; and Han, Y.-S. 2025. TRAPDOC: Deceiving LLM Users by Injecting Imperceptible Phantom Tokens into Documents. arXiv preprint arXiv:2506.00089. ↗
  45. Jobin, A.; Ienca, M.; and Vayena, E. 2019. The global landscape of AI ethics guidelines. Nature machine intelligence, 1(9): 389–399. ↗
  46. Jumelet, J.; Weissweiler, L.; and Bisazza, A. 2025. MultiB-LiMP 1.0: A massively multilingual benchmark of linguistic minimal pairs. arXiv preprint arXiv:2504.02768. ↗
  47. Jurafsky, D.; and Martin, J. H. 2024. Speech and Language Processing. Online draft. 3rd ed. draft, Feb 3, 2024 release. find
  48. Juzek, T. S.; and Ward, Z. B. 2025. Why Does ChatGPT” Delve” So Much? Exploring the Sources of Lexical Overrepresentation in Large Language Models. In Proceedings of the 31st International Conference on Computational Linguistics (COLING 2025). find
  49. Keswani, V.; Conitzer, V.; Heidari, H.; Borg, J. S.; and Sinnott-Armstrong, W. 2024. On the Pros and Cons of Active Learning for Moral Preference Elicitation. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, volume 7, 711–723. find
  50. Kobak, D.; Márquez, R. G.; Horvát, E.-Á.; and Lause, J. 2024. Delving into ChatGPT usage in academic writing through excess vocabulary. arXiv preprint arXiv:2406.07016. ↗
  51. Koppenburg, P. 2024. Tweet on 01 April 2024. https://x.com/PKoppenburg/status/1774757167045788010. Accessed: 2025-01-23. ↗
  52. Kotek, H.; Dockum, R.; and Sun, D. 2023. Gender bias and stereotypes in large language models. In Proceedings of the ACM collective intelligence conference, 12–24. find
  53. Koziev, I. 2025. Detecting Spelling and Grammatical Anomalies in Russian Poetry Texts. arXiv preprint arXiv:2505.04507. ↗
  54. Krielke, M.-P. 2021. Relativizers as markers of grammatical complexity: A diachronic, cross-register study of English and German. Bergen Language and Linguistics Studies, 11(1): 91–120. ↗
  55. Krielke, M.-P. 2023. Optimizing scientific communication: the role of relative clauses as markers of complexity in English and German scientific writing between 1650 and 1900. Saarländische Universitäts-und Landesbibliothek. find
  56. Krielke, M.-P. 2024. Cross-linguistic Dependency Length Minimization in scientific language: Syntactic complexity reduction in English and German in the Late Modern period. Languages in Contrast, 24(1): 133–163. find
  57. Krielke, M.-P.; Talamo, L.; Fawzi, M.; and Knappen, J. 2022. Tracing syntactic change in the scientific genre: Two Universal Dependency-parsed diachronic corpora of scientific English and German. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, 4808– 4816.
  58. Labov, W. 2011. Principles of linguistic change, volume 3: Cognitive and cultural factors, volume 3. John Wiley & Sons. find
  59. Leidinger, A.; and Rogers, R. 2024. How are LLMs mitigating stereotyping harms? Learning from search engine studies. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, volume 7, 839–854. find
  60. Li, Y.; Breithaupt, F.; Hills, T.; Lin, Z.; Chen, Y.; Siew, C. S. W.; and Hertwig, R. 2023. How Cognitive Selection Affects Language Change. Proceedings of the National Academy of Sciences, 121(1). PNAS Direct Submission. Published December 27, 2023. ↗
  61. Liang, W.; Zhang, Y.; Wu, Z.; Lepp, H.; Ji, W.; Zhao, X.; Cao, H.; Liu, S.; He, S.; Huang, Z.; et al. 2024. Mapping the increasing use of llms in scientific papers. arXiv preprint arXiv:2404.01268. ↗
  62. Liu, J.; and Bu, Y. 2024. Towards the relationship between AIGC in manuscript writing and author profiles: evidence from preprints in LLMs. arXiv preprint arXiv:2404.15799. ↗
  63. Lücking, A.; Abrami, G.; Hammerla, L.; Rahn, M.; Baumartz, D.; Eger, S.; and Mehler, A. 2024. Dependencies over times and tools (DoTT). In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), 4641–4653. find
  64. Matsui, K. 2024. Delving into PubMed Records: Some Terms in Medical Writing Have Drastically Changed after the Arrival of ChatGPT. medRxiv, 2024–05. find
  65. Miller, J. E.; and Weinert, R. 1998. Spontaneous spoken language: Syntax and discourse. Oxford University Press. find
  66. Mittelstadt, B. D.; Allo, P.; Taddeo, M.; Wachter, S.; and Floridi, L. 2016. The ethics of algorithms: Mapping the debate. Big Data & Society, 3(2): 2053951716679679. ↗
  67. Montani, I.; Honnibal, M.; Boyd, A.; Landeghem, S. V.; and Peters, H. 2023. explosion/spaCy: v3.7.2: Fixes for APIs and requirements. Version v3.7.2, Zenodo. find
  68. Naik, R.; and Nushi, B. 2023. Social biases through the text-to-image generation lens. In Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society, 786–808. find
  69. Narayanan Venkit, P.; Gautam, S.; Panchanadikar, R.; Huang, T.-H.; and Wilson, S. 2023. Unmasking nationality bias: A study of human perception of nationalities in ai-generated articles. In Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society, 554–565. find
  70. Neff, G.; and Nagy, P. 2016. Talking to bots: Symbiotic agency and the case of Tay. International Journal of Communication, 10: 4915–4931. find
  71. Norhashim, H.; and Hahn, J. 2024. Measuring Human-AI Value Alignment in Large Language Models. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, volume 7, 1063–1073. find
  72. Omiye, J. A.; Lester, J. C.; Spichak, S.; Rotemberg, V.; and Daneshjou, R. 2023. Large language models propagate race-based medicine. NPJ Digital Medicine, 6(1): 195. ↗
  73. Omrani Sabbaghi, S.; Wolfe, R.; and Caliskan, A. 2023. Evaluating biased attitude associations of language models in an intersectional context. In Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society, 542–553. find
  74. OpenAI. 2024. Tweet on 08 April 2024. https://x.com/ChatGPTapp/status/1777221658807521695. Accessed: 2024-08-12. ↗
  75. Ouyang, L.; Wu, J.; Jiang, X.; Almeida, D.; Wainwright, C.; Mishkin, P.; Zhang, C.; Agarwal, S.; Slama, K.; Ray, A.; et al. 2022. Training language models to follow instructions with human feedback. Advances in neural information processing systems, 35: 27730–27744.
  76. Pack, A.; and Maloney, J. 2023. Potential Affordances of Generative AI in Language Education: Demonstrations and an Evaluative Framework. Teaching English with Technology, 23(2). ↗
  77. Perez, E.; Ringer, S.; Lukosiute, K.; Nguyen, K.; Chen, E.; Heiner, S.; Pettit, C.; Olsson, C.; Kundu, S.; Kadavath, S.; et al. 2023. Discovering language model behaviors with model-written evaluations. In Findings of the Association for Computational Linguistics: ACL 2023, 13387–13434. ↗
  78. Python Software Foundation. 2024. Python 3. find
  79. Radford, A.; Kim, J. W.; Xu, T.; Brockman, G.; McLeavey, C.; and Sutskever, I. 2022. Whisper. https://github.com/ openai/whisper. ↗
  80. Rafailov, R.; Sharma, A.; Mitchell, E.; Manning, C. D.; Ermon, S.; and Finn, C. 2024. Direct preference optimization: Your language model is secretly a reward model. Advances in Neural Information Processing Systems, 36. ↗
  81. Russell, S.; Dewey, D.; and Tegmark, M. 2015. Research priorities for robust and beneficial artificial intelligence. AI magazine, 36(4): 105–114. ↗
  82. Sadasivan, V. S.; Kumar, A.; Balasubramanian, S.; Wang, W.; and Feizi, S. 2023. Can AI-generated text be reliably detected? arXiv preprint arXiv:2303.11156. ↗
  83. Sandra, G. 2011. 7. Greiffenstern, Sandra. 2010. The influence of computers, the internet and computer-mediated communication on everyday English. English and American Studies in German, 2011(1): 16–18. find
  84. Schmalz, V.; and Tack, A. 2025. Can GPTZero’s AI Vocabulary Distinguish Between LLM-Generated and Student-Written Essays? In Kochmar, E.; Alhafni, B.; Bexte, M.; Burstein, J.; Horbach, A.; Laarmann-Quante, R.; Tack, A.; Yaneva, V.; and Yuan, Z., eds., Proceedings of the 20th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2025), 937–952. Vienna, Austria: Association for Computational Linguistics. ISBN 979-8-89176-270-1. ↗
  85. Shumailov, I.; Shumaylov, Z.; Zhao, Y.; Gal, Y.; Papernot, N.; and Anderson, R. 2023. The curse of recursion: Training on generated data makes models forget. arXiv preprint arXiv:2305.17493. ↗
  86. Srivastava, A.; Rastogi, A.; Rao, A.; Shoeb, A. A.; Abid, A.; Fisch, A.; Brown, A. R.; Santoro, A.; Gupta, A.; Garriga-Alonso, A.; et al. 2023. Beyond the imitation game: Quantifying and extrapolating the capabilities of language models. Transactions on machine learning research.
  87. Tomasello, M. 2005. Constructing a language: A usage-based theory of language acquisition. Harvard university press. find
  88. Traugott, E. C. 1985. On regularity in semantic change. Journal of Literary Semantics, 14(3): 155–173. ↗
  89. Vanisree, M.; Sharma, S.; Singh, L.; Kantharaja, K. P.; and Sri Nidhya, G. 2024. Effects of Technology and Social Media on English Language Use and Acquisition: A Study on Language Variation and Change. Nanotechnology Perceptions, 20(S7): 490–497. ↗
  90. Wang, G.; Wang, H.; Sun, X.; Wang, N.; and Wang, L. 2023. Linguistic complexity in scientific writing: A large-scale diachronic study from 1821 to 1920. Scientometrics, 128(1): 441–460. find
  91. Warstadt, A.; Parrish, A.; Liu, H.; Mohananey, A.; Peng, W.; Wang, S.-F.; and Bowman, S. R. 2020. BLiMP: The benchmark of linguistic minimal pairs for English. Transactions of the Association for Computational Linguistics, 8: 377–392. ↗
  92. Weber-Wulff, D.; Anohina-Naumeca, A.; Bjelobaba, S.; Foltỳnek, T.; Guerrero-Dib, J.; Popoola, O.; Šigut, P.; and Waddington, L. 2023. Testing of detection tools for AI-generated text. International Journal for Educational Integrity, 19(1): 1–39. ↗
  93. Weissweiler, L.; Mahowald, K.; and Goldberg, A. 2025. Linguistic generalizations are not rules: Impacts on evaluation of LMs. arXiv preprint arXiv:2502.13195. ↗
  94. Welbl, J.; Glaese, A.; Uesato, J.; Dathathri, S.; Mellor, J.; Hendricks, L. A.; Anderson, K.; Kohli, P.; Coppin, B.; and Huang, P.-S. 2021. Challenges in detoxifying language models. arXiv preprint arXiv:2109.07445. ↗
  95. Wilson, K.; and Caliskan, A. 2024. Gender, race, and intersectional bias in resume screening via language model retrieval. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, volume 7, 1578–1590. find
  96. Wu, M.; and Aji, A. F. 2025. Style Over Substance: Evaluation Biases for Large Language Models. In Rambow, O.; Wanner, L.; Apidianaki, M.; Al-Khalifa, H.; Eugenio, B. D.; and Schockaert, S., eds., Proceedings of the 31st International Conference on Computational Linguistics, 297–312. Abu Dhabi, UAE: Association for Computational Linguistics.
  97. Wu, Y.; Bizzoni, Y.; Moreira, P.; and Nielbo, K. 2024. Perplexing canon: A study on GPT-based perplexity of canonical and non-canonical literary works. In Proceedings of the 8th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (LaTeCH-CLfL 2024), 172–184. find
  98. Yakura, H.; Lopez-Lopez, E.; Brinkmann, L.; Serna, I.; Gupta, P.; and Rahwan, I. 2024. Empirical evidence of Large Language Model’s influence on human spoken communication. arXiv preprint arXiv:2409.01754. ↗
  99. Zamaraeva, O.; Flickinger, D.; Bond, F.; and Gómez-Rodrı́guez, C. 2025. Comparing LLM-generated and human-authored news text using formal syntactic theory. arXiv preprint arXiv:2506.01407. ↗
  100. Zhang, R.; Ouni, J.; and Eger, S. 2024. Cross-lingual cross-temporal summarization: Dataset, models, evaluation. Computational Linguistics, 50(3): 1001–1047. ↗
  101. Zhang, X.; Xiong, W.; Chen, L.; Zhou, T.; Huang, H.; and Zhang, T. 2024. From lists to emojis: How format bias affects model alignment. arXiv preprint arXiv:2409.11704. ↗
  102. Zhang, Y. 2014. A Review of Principles of Linguistic Change: Cognitive and Cultural Factors. Open Journal of Modern Linguistics, 4: 65–68. ↗
  103. Ziegler, D. M.; Stiennon, N.; Wu, J.; Brown, T. B.; Radford, A.; Amodei, D.; Christiano, P.; and Irving, G. 2019. Fine-tuning language models from human preferences. arXiv preprint arXiv:1909.08593. Word irreplaceable ADJ eminent ADJ territory NOUN exhibit NOUN admirable ADJ capability NOUN multifaceted ADJ probable ADJ notable ADJ achievable ADJ assemble VERB thorough ADJ reinforce VERB novel ADJ grasp VERB complexity NOUN visionary ADJ enhancement NOUN prospect NOUN stress VERB integrate VERB sophisticated ADJ coordinate VERB elaborate ADJ deliberately ADV remarkable ADJ critical ADJ vital ADJ original ADJ highlight VERB improvement NOUN complex ADJ vigilant ADJ showcase VERB convoluted ADJ possibility NOUN innovative ADJ merit NOUN complication NOUN renowned ADJ pivotal ADJ inventive ADJ ↗

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/anderson-2025-model-misalignment-language-change-traces/