Title: CURP: Codebook-based Continuous User Representation for Personalized Generation with LLMs

Original: arXiv:2602.00742v2 Announce Type: replace Abstract: User modeling characterizes individuals through their preferences and behavioral patterns to enable personalized simulation and generation with Large Language Models (LLMs) in contemporary approaches. However, existing methods, whether prompt-based or training-based methods, face challenges in balancing personalization quality against computational and data efficiency. We propose a novel framework CURP, which employs a bidirectional user encoder and a discrete prototype codebook to extract multi-dimensional user traits. This design enables plug-and-play personalization with a small number of trainable parameters (about 20M parameters, about 0.2\% of the total model size). Through extensive experiments on variant generation tasks, we show that CURP achieves superior performance and generalization compared to strong baselines, while offering better interpretability and scalability. The code are available at https://github.com/RaidonWong/CURP_code

Rewrite: Current strategies for personalizing Large Language Model (LLM) simulations and content generation rely on user modeling, which defines individuals based on their behaviors and preferences. Nevertheless, these existing approaches—whether they depend on prompt engineering or model training—struggle to strike an effective balance between the quality of personalization and the demands placed on computational resources and data. To address this, we introduce CURP, a new framework that utilizes a discrete prototype codebook alongside a bidirectional user encoder to identify multi-dimensional user characteristics. This architecture facilitates immediate, plug-and-play personalization while requiring minimal trainable parameters, specifically around 20 million, which constitutes merely 0.2% of the overall model capacity. Our comprehensive evaluations across various generation tasks demonstrate that CURP outperforms robust baseline methods in both generalization and performance, while also providing enhanced interpretability and scalability. The source code can be accessed at https://github.com/RaidonWong/CURP_code

Source: arXiv Generated at: 2026-06-02 00:00:00 UTC