ACL Findings2026
Chameleons and Guardians: Unveiling the Divergence in Personality Plasticity and Cognitive Resistance across LLMs
Jiaxin Wu, Jie Zhao, Xinrong Huang, Cuihong Zhang, Xiuzhu Wu
Abstract
Whether the personality of LLMs can be intentionally reshaped remains controversial.Existing studies often limited to small models, argue for its immutability.Crucially, prior studies fail to uncover that different LLMs exhibit significant compliance divergence when exposed to personality-inducing contexts.To bridge this gap, we introduce Personality Induction Framework (PIF), which systematically reshapes the personality of different LLMs via multi-agent collaboration.Specifically, via Generator-Judge agents, PIF paraphrases MBTI questions to create semantically equivalent but expressively diverse inducing contexts, enabling LLMs to learn personality patterns instead of superficial token matching.Also, PIF achieves fine-grained personality modulation by controlling the intensity of inducing contexts.Extensive experiments on worldwide mainstream LLMs show that PIF reliably transforms their original personalities into desired target personalities.Notably, we find that the outputs of most Western LLMs behave like "Chameleons", exhibiting high personality plasticity; whereas the outputs of most Eastern LLMs act as "Guardians", manifesting pronounced cognitive resistance.Strikingly, extreme induction intensity (100%) triggers a counter-intuitive "Alignment Rebound" in Guardians, resulting in the opposite direction rather than compliance.These findings suggest that LLM personality is a dynamic equilibrium shaped by the trade-off between instruction compliance and cognitive resistance.G e m in i3 _ 0 .0
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on25
- RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language ModelsNoah Wang, Zhongyuan Peng, Haoran Que, Jiaheng Liu et al.ACL Findings 2024 · 120 citations
- Character-LLM: A Trainable Agent for Role-PlayingYunfan Shao, Linyang Li, Junqi Dai, Xipeng QiuEMNLP 2023 · 97 citations
- Two Tales of Persona in LLMs: A Survey of Role-Playing and PersonalizationYu-Min Tseng, Yu-Chao Huang, Teng-Yun Hsiao, Wei-Lin Chen et al.EMNLP Findings 2024 · 87 citations
- MuMA-ToM: Multi-modal Multi-Agent Theory of MindHaojun Shi, Suyu Ye, Xinyu Fang, Chuanyang Jin et al.AAAI 2025 · 48 citations
- Open Models, Closed Minds? On Agents Capabilities in Mimicking Human Personalities through Open Large Language ModelsLucio La Cava, Andrea TagarelliAAAI 2025 · 36 citations
Related papers
- Can LLM Agents Maintain a Persona in Discourse?Pranav Bhandari, Nicolas Fay, Michael J. Wise, Amitava Datta et al.EMNLP 2025
- The Personality Illusion: Revealing Dissociation Between Self-Reports & Behavior in LLMsPengrui Han, Rafal Kocielnik, Peiyang Song, Ramit Debnath et al.ICML 2026 · 33 citations
- Chameleon LLMs: User Personas Influence Chatbot Personality ShiftsJane Xing, Tianyi Niu, Shashank SrivastavaEMNLP 2025
- From Single to Societal: Analyzing Persona-Induced Bias in Multi-Agent InteractionsJiayi Li, Xiao Liu, Yansong FengAAAI 2026 · 3 citations
- Vibe Check: Understanding the Effects of LLM-Based Conversational Agents' Personality and Alignment on User Perceptions in Goal-Oriented TasksHasibur Rahman, Smit DesaiCHI 2026 · 1 citation
