PersonaEdit: Representative Sample Selection for Personalized Model Editing
Computation and Language
Summary
The authors studied how to personalize large language models (LLMs) by directly changing their internal settings, a process called model editing. They found that editing many user data points is costly and can cause conflicts, so they created PersonaEdit, a method that picks only important examples using a smart sampling approach. Their tests showed this approach keeps most of the performance while using fewer samples. They also discovered that combining model editing with traditional retrieval methods works even better for personalization.
Authors
You-Mei Huang, Chung-Chi Chen, An-Zi Yen
Abstract
Personalization has attracted growing interest in LLM applications, yet existing retrieval-based approaches depend heavily on retrieval quality and degrade in long-term interactions. Model editing, which directly modifies internal model parameters to incorporate new knowledge, has demonstrated effective knowledge modification capabilities in factual knowledge editing tasks and may provide a potential solution for personalization. However, scaling model editing to personalization is non-trivial. Editing large amounts of user data increases computational cost and causes interference among edits, motivating the need for effective sample selection. To address this issue, we propose, PersonaEdit, a hidden representation clustering strategy that selects representative editing samples through proportional stratified sampling. Experiments show that model editing is effective for personalization, and that our selection strategy preserves most of the performance while substantially reducing the number of required editing samples. Beyond standalone editing, we find that combining model editing with retrieval-based prompt augmentation further improves personalization, as edited knowledge and retrieved context provide complementary information. These results demonstrate the potential of model editing as an efficient and scalable approach for LLM personalization.