Evaluating LLMs for Impersonation

N/A

Preprint IMPersona: Evaluating Individual Level LM Impersonation Quan Shi, Carlos E. Jimenez, Stephen Dong, Brian Seo, Caden Yao, Adam Kelch, Karthik Narasimhan Princeton Language and Intelligence (PLI), Princeton University {qbshi, carlosej}@princeton.edu Abstract arXiv:2504.04332v2…

related reading

• [2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
• Prompt Injection as Role Confusionrole-confusion.github.io
• [2606.06614] Re-Centering Humans in LLM Personalizationarxiv.org
• [2512.11949] Neural Chameleons: Language Models Can Learn to Hide Their Thoughts from Unseen Activation Monitorsarxiv.org
• Prefill awareness: can LLMs tell when “their” message history has been tampered with? — LessWronglesswrong.com
• Discovering Language Model Behaviors with Model-Written Evaluations — LessWronglesswrong.com
• Training and Evaluating User Language Modelsarxiv.org
• [2502.05209] Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilitiesarxiv.org
• [2607.14111] Introspection Fine-Tuning (IFT): Training Small LLMs to Introspectarxiv.org
• Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
• Language Models can Solve Computer Tasksarxiv.org
• [2412.00543] Evaluating the Consistency of LLM Evaluatorsarxiv.org