Elamur Research⌕
☰
home / papers / 2026 / Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models

Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models

Paper · 2026 · Arya Shah, Deepali Mishra and Chaklam Silpasuwanchai

Summary

Chat models are often asked to play characters, and this raises a question: does the personality of the assigned character make the model more likely to tell users what they want to hear? The authors test whether persona agreeableness, the Big Five trait linked to cooperation and avoiding conflict, predicts sycophancy. They built 275 personas, ranging from abrasive to warm, and scored each one by having the model answer an adapted NEO-IPIP agreeableness questionnaire. Each persona then faced 4,950 opinionated "don't you agree?" prompts across 33 topic categories. The study covered 13 open-weight models from 0.6B to 20B parameters, with stances detected automatically. Two metrics compare persona behaviour with a plain-assistant baseline: SSIP and the Trait-Truthfulness Gap (TTG). Within most models, more agreeable personas agreed with users more often. However, assigning a persona usually lowered sycophancy compared with the baseline. This shows that persona traits matter for safety, and the authors argue that agreeable characters in deployed systems need explicit truthfulness safeguards.

Key points

  • 9 of the 13 models show a statistically significant positive correlation between agreeableness and sycophancy (α = 0.05). The strongest are Llama 3.1 8B (r = 0.868, R² = 0.753) and OLMo 3 7B (r = 0.853).
  • The paper reports effect sizes inconsistently. The abstract cites Cohen's d up to 2.33 (SmolLM3 3B), while the conclusion gives a range of 0.455 to 1.282.
  • Mean TTG is negative for most models, meaning personas made them more truthful than baseline. Llama 3.1 8B has the lowest value (−0.43), with 99.3% of its personas in the "truthful zone."
  • Gemma 3 1B is the exception. Its mean TTG is +0.34, and 94.9% of its personas fall in the "zone of deception."
  • The evaluation totals about 17.9M queries. Code is released on GitHub and the dataset on Hugging Face. Stated limitations include automated stance detection, subjective single-turn prompts only, and a focus on agreeableness alone.

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

🎓Indian Institute of Technology KanpurKanpur · India14 works · 619 citations🎓Indian Institute of Technology GandhinagarGandhinagar · India2 works · 2 citations

Also on the byline: Asian Institute of Technology.

 Fields, topics and concepts

📚 Shelves 6The 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 worksSocial Sciences · 1,772 worksSafety Research · 543 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 3,320 works

Cite

Cite as Arya Shah, Deepali Mishra and Chaklam Silpasuwanchai. 2026. Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.1421

@inproceedings{shah2026too,
  title = {{Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models}},
  author = {Arya Shah, Deepali Mishra and Chaklam Silpasuwanchai},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1421},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1421}
}

References

52 references, 21 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. 01. AI, Alex Young, Bei Chen, Chao Li, Chengen Huang, Ge Zhang, Guanwei Zhang, Guoyin Wang, Heng Li, Jiangcheng Zhu, Jianqun Chen, Jing Chang, Kaidong Yu, Peng Liu, Qiang Liu, Shawn Yue, Senbin Yang, Shiming Yang, Wen Xie, and 13 others. 2025. Yi: Open foundation models by 01.ai. Preprint, arXiv:240…
  2. Alexander Amini, Anna Banaszak, Harold Benoit, Arthur Böök, Tarek Dakhran, Song Duong, Alfred Eng, Fernando Fernandes, Marc Härkönen, Anne Harrington, Ramin Hasani, Saniya Karwa, Yuri Khrustalev, Maxime Labonne, Mathias Lechner, Valentine Lechner, Simon Lee, Zetian Li, Noel Loo, and 14 others. 2025.… — [arXiv:2511.23404](https://arxiv.org/abs/2511.23404) find
  3. Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan, Nicholas Joseph, Saurav Kadavath, Jackson Kernion, Tom Conerly, Sheer El-Showk, Nelson Elhage, Zac Hatfield-Dodds, Danny Hernandez, Tristan Hume, and 12 others. 202… find
  4. Elie Bakouch, Loubna Ben Allal, Anton Lozhkov, Nouamane Tazi, Lewis Tunstall, Carlos Miguel Patiño, Edward Beeching, Aymeric Roucher, Aksel Joonas Reedi, Quentin Gallouédec, Kashif Rasul, Nathan Habib, Clémentine Fourrier, Hynek Kydlicek, Guilherme Penedo, Hugo Larcher, Mathieu Morlon, Vaibhav Sriva… find
  5. Yassine El Boudouri, Walter Nuninger, Julian Alvarez, and Yvan Peter. 2025. Role-playing evaluation for large language models. Preprint, arXiv:2505.13157. — [arXiv:2505.13157](https://arxiv.org/abs/2505.13157) find
  6. Dallas Card, Peter Henderson, Urvashi Khandelwal, Robin Jia, Kyle Mahowald, and Dan Jurafsky. 2020. With little power comes great responsibility. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 9263–9274, Online. Association for Computational… find
  7. Chien Hung Chen, Hen-Hsen Huang, and Hsin-Hsi Chen. 2025. Self-augmented preference alignment for sycophancy reduction in LLMs. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 12390–12402, Suzhou, China. Association for Computational Linguistics. find
  8. Jiangjie Chen, Xintao Wang, Rui Xu, Siyu Yuan, Yikai Zhang, Wei Shi, Jian Xie, Shuang Li, Ruihan Yang, Tinghui Zhu, Aili Chen, Nianqi Li, Lida Chen, Caiyu Hu, Siye Wu, Scott Ren, Ziquan Fu, and Yanghua Xiao. 2024. From persona to personalization: A survey on role-playing language agents. Preprint, a…
  9. Aileen Cheng, Alon Jacovi, Amir Globerson, Ben Golan, Charles Kwong, Chris Alberti, Connie Tao, Eyal Ben-David, Gaurav Singh Tomar, Lukas Haas, Yonatan Bitton, Adam Bloniarz, Aijun Bai, Andrew Wang, Anfal Siddiqui, Arturo Bajuelos Castillo, Aviel Atias, Chang Liu, Corey Fry, and 46 others. 2025a. Th… — [arXiv:2512.10791](https://arxiv.org/abs/2512.10791) find
  10. Myra Cheng, Sunny Yu, Cinoo Lee, Pranav Khadpe, Lujain Ibrahim, and Dan Jurafsky. 2025b. Elephant: Measuring and understanding social sycophancy in llms. Preprint, arXiv:2505.13995.
  11. Jacob Cohen. 1992. Statistical power analysis. Current Directions in Psychological Science, 1(3):98–101. find
  12. Paul T Costa and Robert R McCrae. 2008. The revised NEO personality inventory (NEO-PI-R). In The SAGE Handbook of Personality Theory and Assessment: Volume 2 — Personality Measurement and Testing, pages 179–198. SAGE Publications Ltd, 1 Oliver’s Yard, 55 City Road, London EC1Y 1SP United Kingdom. find
  13. Pegah Jandaghi, Xianghai Sheng, Xinyi Bai, Jay Pujara, and Hakim Sidahmed. 2024. Faithful persona-based conversational dataset generation with large language models. In Proceedings of the 6th Workshop on NLP for Conversational AI (NLP4ConvAI 2024), pages 114–139, Bangkok, Thailand. Association for C… find
  14. Rotem Dror, Gili Baumer, Segev Shlomov, and Roi Reichart. 2018. The hitchhiker’s guide to testing statistical significance in natural language processing. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1383–1392, Melbourne, A… find
  15. Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, Lélio Renard Lavaud, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix,… find
  16. Tim Duffy. 2025. Syco-bench: A multi-part benchmark for sycophancy in LLMs. https://www.syco-bench.com/syco-bench.pdf. Code available at https://github.com/ timfduffy/syco-bench. ↗
  17. Aaron Fanous, Jacob Goldberg, Ank A. Agarwal, Joanna Lin, Anson Zhou, Roxana Daneshjou, and Sanmi Koyejo. 2025. Syceval: Evaluating llm sycophancy. Preprint, arXiv:2502.08177. — [arXiv:2502.08177](https://arxiv.org/abs/2502.08177) find
  18. Tao Ge, Xin Chan, Xiaoyang Wang, Dian Yu, Haitao Mi, and Dong Yu. 2025. Scaling synthetic data creation with 1,000,000,000 personas. Preprint, arXiv:2406.20094.
  19. Lewis R Goldberg and 1 others. 1999. A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several five-factor models. Personality psychology in Europe, 7(1):7– 28. find
  20. Granite Team, IBM. 2025. Granite-3.3-8b-instruct. Hugging Face Model Repository. Release date: April 16, 2025. find
  21. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sravankumar, Artem Korenev, Arthur Hinsvark, and 542 o… — The Llama 3 Herd of Models
  22. Hang Jiang, Xiajie Zhang, Xubo Cao, Cynthia Breazeal, Deb Roy, and Jad Kabbara. 2024. PersonaLLM: Investigating the ability of large language models to express personality traits. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 3605–3627, Mexico City, Mexico. Associat… find
  23. Haibo Jin, Ruoxi Chen, Peiyan Zhang, Andy Zhou, and Haohan Wang. 2025. Guard: Guideline upholding test through adaptive role-play and jailbreak diagnostics for llms. Preprint, arXiv:2508.20325. — [arXiv:2508.20325](https://arxiv.org/abs/2508.20325) find
  24. Junyi Li, Xiaoxue Cheng, Wayne Xin Zhao, Jian-Yun Nie, and Ji-Rong Wen. 2023. Halueval: A large-scale hallucination evaluation benchmark for large language models. Preprint, arXiv:2305.11747.
  25. Stephanie Lin, Jacob Hilton, and Owain Evans. 2022. TruthfulQA: Measuring how models mimic human falsehoods. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3214–3252, Dublin, Ireland. Association for Computational Linguistics…
  26. Microsoft, Abdelrahman Abouelenin, Atabak Ashfaq, Adam Atkinson, Hany Awadalla, Nguyen Bach, Jianmin Bao, Alon Benhaim, Martin Cai, Vishrav Chaudhary, Congcong Chen, Dong Chen, Dongdong Chen, Junkun Chen, Weizhu Chen, Yen-Chun Chen, Yi ling Chen, Qi Dai, Xiyang Dai, and 56 others. 2025. Phi-4-mini t…
  27. Jiseung Hong, Grace Byun, Seungone Kim, Kai Shu, and Jinho D. Choi. 2025. Measuring sycophancy of language models in multi-turn dialogues. Preprint, arXiv:2505.23840. find
  28. Dor Muhlgay, Ori Ram, Inbal Magar, Yoav Levine, Nir Ratner, Yonatan Belinkov, Omri Abend, Kevin Leyton-Brown, Amnon Shashua, and Yoav Shoham. 2024. Generating benchmarks for factuality evaluation of language models. In Proceedings of the 18th Conference of the European Chapter of the Association for…
  29. Evan Hubinger. 2023. Modulating sycophancy in an RLHF model via activation steering. AI Alignment Forum. Accessed: December 30, 2025. find
  30. Team Olmo, Allyson Ettinger, Amanda Bertsch, Bailey Kuehl, David Graham, David Heineman, Dirk Groeneveld, Faeze Brahman, Finbarr Timbers, Hamish William G Graziano and Nancy Eisenberg. 1997. Agreeableness. In Handbook of Personality Psychology, pages 795–824. Elsevier. find
  31. Ivison, Jacob Morrison, Jake Poznanski, Kyle Lo, Luca Soldaini, Matt Jordan, Mayee Chen, Michael Noukhovitch, Nathan Lambert, Pete Walsh, and 49 others. 2025. Olmo 3. Preprint, arXiv:2512.13961.
  32. Oliver Rausch, Nicholas Schiefer, Da Yan, Miranda Zhang, and Ethan Perez. 2025. Towards understanding sycophancy in language models. Preprint, arXiv:2310.13548.
  33. OpenAI, Sandhini Agarwal, Lama Ahmad, Jason Ai, Sam Altman, Andy Applebaum, Edwin Arbus, Rahul K. Arora, Yu Bai, Bowen Baker, Haiming Bao, Boaz Barak, Ally Bennett, Tyler Bertao, Nivedita Brett, Eugene Brevdo, Greg Brockman, Sebastien Bubeck, Che Chang, and 107 others. 2025. gpt-oss-120b and gpt-oss…
  34. Tom Sühr, Florian E. Dorner, Samira Samadi, and Augustin Kelava. 2024. Challenging the validity of personality tests for large language models. Preprint, arXiv:2311.05297. — [arXiv:2311.05297](https://arxiv.org/abs/2311.05297) find
  35. Yihong Tang, Kehai Chen, Xuefeng Bai, Zhengyu Niu, Bo Wang, Jie Liu, and Min Zhang. 2025. The rise of darkness: Safety-utility trade-offs in role-playing dialogue agents. Preprint, arXiv:2502.20757. find
  36. Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. 2022. Train…
  37. Gemma Team, Aishwarya Kamath, Johan Ferret, Shreya Pathak, Nino Vieillard, Ramona Merhej, Sarah Perrin, Tatiana Matejovicova, Alexandre Ramé, Morgane Rivière, Louis Rouillard, Thomas Mesnard, Geoffrey Cideron, Jean bastien Grill, Sabela Ramos, Edouard Yvinec, Michelle Casbon, Etienne Pot, Ivo Penche… find
  38. Ethan Perez, Sam Ringer, Kamilė Lukošiūtė, Karina Nguyen, Edwin Chen, Scott Heiner, Craig Pettit, Catherine Olsson, Sandipan Kundu, Saurav Kadavath, Andy Jones, Anna Chen, Ben Mann, Brian Israel, Bryan Seethor, Cameron McKinnon, Christopher Olah, Da Yan, Daniela Amodei, and 44 others. 2022. Disco…
  39. MiniCPM Team, Chaojun Xiao, Yuxuan Li, Xu Han, Yuzhuo Bai, Jie Cai, Haotian Chen, Wentong Chen, Xin Cong, Ganqu Cui, Ning Ding, Shengda Fan, Yewei Fang, Zixuan Fu, Wenyu Guan, Yitong Guan, Junshao Guo, Yufeng Han, Bingxiang He, and 64 others. 2025b. Minicpm4: Ultra-efficient llms on end devices. Pre…
  40. Ivo Petrov, Jasper Dekoninck, and Martin Vechev. 2025. Brokenmath: A benchmark for sycophancy in theorem proving with llms. Preprint, arXiv:2510.04721. find
  41. Qwen, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, and 24 others. 2025. Qwen2.5 technical report. Preprint, arXiv:2412.15… find
  42. Greg Serapio-García, Mustafa Safdari, Clément Crepy, Luning Sun, Stephen Fitz, Peter Romero, Marwa Abdulhai, Aleksandra Faust, and Maja Matarić. 2025. Personality traits in large language models. Preprint, arXiv:2307.00184.
  43. Rusheb Shah, Quentin Feuillade-Montixi, Soroush Pour, Arush Tagade, Stephen Casper, and Javier Rando. 2023. Scalable and transferable black-box jailbreaks for language models via persona modulation. Preprint, arXiv:2311.03348. — [arXiv:2311.03348](https://arxiv.org/abs/2311.03348) find
  44. Murray Shanahan, Kyle McDonell, and Laria Reynolds. 2023. Role-play with large language models. Preprint, arXiv:2305.16367.
  45. Mrinank Sharma, Meg Tong, Tomasz Korbak, David Duvenaud, Amanda Askell, Samuel R. Bowman, Newton Cheng, Esin Durmus, Zac Hatfield-Dodds, Scott R. Johnston, Shauna Kravec, Timothy Maxwell, Sam McCandlish, Kamal Ndousse, Tommaso Tosato, Saskia Helbling, Yorguin-Jose Mantilla-Ramos, Mahmood Hegazy, Alb… — [arXiv:2508.04826](https://arxiv.org/abs/2508.04826) find
  46. Quan Tu, Shilong Fan, Zihang Tian, Tianhao Shen, Shuo Shang, Xin Gao, and Rui Yan. 2024. CharacterEval: A Chinese benchmark for role-playing conversational agent evaluation. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1183…
  47. Lei Wang, Jianxun Lian, Yi Huang, Yanqi Dai, Haoxuan Li, Xu Chen, Xing Xie, and Ji-Rong Wen. 2024. Characterbox: Evaluating the role-playing capabilities of llms in text-based virtual worlds. Preprint, arXiv:2412.05631. find
  48. Jerry Wei, Da Huang, Yifeng Lu, Denny Zhou, and Quoc V. Le. 2024. Simple synthetic data reduces sycophancy in large language models. Preprint, arXiv:2308.03958. find
  49. Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pier-ric Cistac, Tim Rault, Rémi Louf, Morgan Funtowicz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, and 3 others. 2020. Huggin…
  50. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, Chujie Zheng, Dayiheng Liu, Fan Zhou, Fei Huang, Feng Hu, Hao Ge, Haoran Wei, Huan Lin, Jialong Tang, and 41 others. 2025. Qwen3 technical report. Preprint, arXiv:2505.09388.
  51. Baohua Zhan, Yongyi Huang, Wenyao Cui, Huaping Zhang, and Jianyun Shang. 2024. Humanity in ai: Detecting the personality of large language models. Preprint, arXiv:2410.08545. — [arXiv:2410.08545](https://arxiv.org/abs/2410.08545) find
  52. Jinman Zhao, Zifan Qian, Linbo Cao, Yining Wang, Yitian Ding, Yulan Hu, Zeyu Zhang, and Zeyong Jin. 2025. Role-play paradox in large language models: Reasoning performance gains and ethical dilemmas. Preprint, arXiv:2409.13979.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/shah-2026-too-nice-tell-truth-quantifying/