Large language models can forget specific personas without losing skill

LLM Persona Unlearning

Computation and LanguageMachine Learning

Summary

Large language models learn many ways to act, like different personalities, but they can still show unwanted behaviors even after training. The authors spot a problem: current methods can't reliably erase these unwanted personas without harming the model's usefulness. They created a new way called PaCE that helps the model 'forget' specific personas while keeping good responses to other questions. This approach shows promise to keep AI assistants behaving consistently and avoids unwanted personality traits.

What this means in practice

  • For language model developers: Improve model training by enabling permanent removal of unwanted personas while preserving general capabilities.
  • For chatbot engineers: Enable chatbots to avoid undesirable behavior traits through targeted persona suppression in model weights.

Authors

Kemou Li, Zhuan Shi, Qizhou Wang, Fengpeng Li, Negar Rostamzadeh, Golnoosh Farnadi, Jiantao Zhou

Abstract

Pre-training equips large language models (LLMs) with a broad repertoire of behavioral patterns associated with roles, styles, values, and goals. Post-training teaches conditional enactment and makes a helpful Assistant the default, but it does not erase alternative modes from the weights; explicit prompts can therefore elicit personas that repeatedly shape judgment, language, and action. In open-weight settings, runtime controls can be removed, motivating persona unlearning: a weight-level edit that makes a designated persona difficult to elicit and enact on unseen contexts. We introduce PersonaUnlearnBench, a model-specific paired benchmark spanning six LLMs from three families and five personas, with aligned forget/retain sets, held-out instruction paraphrases, and four-axis evaluation. The benchmark shows that standard unlearning methods cannot reliably erase the target persona without sacrificing meaningful generation or general utility. We therefore propose PaCE, which compares target and desirable responses to the same questions to locate an internal behavior direction, then trains target-prompt states away from the target mode and toward the matched desirable response. Experiments show that PaCE consistently suppresses target personas with high response quality and useful counterpart behavior, at moderate utility cost. These results establish persona unlearning as a distinct behavior-level editing problem and a practical route toward persistent control of latent LLM response policies.