PMG : Personalized Multimodal Generation with Large Language Models

TL;DR

This paper introduces PMG, a novel method for personalized multimodal content generation using large language models, which effectively incorporates user preferences to improve personalization without sacrificing accuracy.

Contribution

It presents the first approach to personalized multimodal generation with LLMs, converting user behaviors into preferences to guide content creation.

Findings

PMG improves personalization by up to 8% in LPIPS scores.

The method effectively balances accuracy and personalization.

Experimental validation on two datasets demonstrates its effectiveness.

Abstract

The emergence of large language models (LLMs) has revolutionized the capabilities of text comprehension and generation. Multi-modal generation attracts great attention from both the industry and academia, but there is little work on personalized generation, which has important applications such as recommender systems. This paper proposes the first method for personalized multimodal generation using LLMs, showcases its applications and validates its performance via an extensive experimental study on two datasets. The proposed method, Personalized Multimodal Generation (PMG for short) first converts user behaviors (e.g., clicks in recommender systems or conversations with a virtual assistant) into natural language to facilitate LLM understanding and extract user preference descriptions. Such user preferences are then fed into a generator, such as a multimodal LLM or diffusion model, to produce personalized content. To capture user preferences comprehensively and accurately, we propose to let the LLM output a combination of explicit keywords and implicit embeddings to represent user preferences. Then the combination of keywords and embeddings are used as prompts to condition the generator. We optimize a weighted sum of the accuracy and preference scores so that the generated content has a good balance between them. Compared to a baseline method without personalization, PMG has a significant improvement on personalization for up to 8% in terms of LPIPS while retaining the accuracy of generation.