Personalized Image Generation with Reasoning and Reflection

2026-09-30

AI Summary

This study addresses the limitation of conventional personalized image generation, which relies solely on sparse visual exemplars and fails to exploit users’ rich historical contextual information. To overcome this, we propose PEARL, a novel framework that introduces the first user-history benchmark incorporating multi-axis evaluation. Furthermore, we design an interleaved reasoning-and-reflection optimization mechanism that integrates a multimodal reasoner, a frozen image generator, and differential data reward optimization to enable personalized generation for both scene-consistent and creative imagery. As the first unified benchmark and generation framework grounded in user history, PEARL outperforms strong baselines across two tasks, achieving an average improvement of 15% in personalization metrics.

Abstract

Personalized image generation has remained narrowly focused on conditional synthesis from curated visual exemplars, rather than capturing who a user is. In practice, however, a user's personal context is much richer, comprising reviews, posts, images, captions, and metadata accumulated over time. A truly personalized generator should leverage this history to produce images aligned with the user's lifestyle and aesthetic preferences. To this end, we introduce the first unified benchmark for personalized image generation from user histories. The benchmark comprises two complementary tasks and a multi-axis evaluation protocol that assesses target fidelity, visual quality, user distinguishability, semantic alignment with the user's history, and task-specific utility. Grounded in real-world e-commerce and social media settings, the benchmark includes: (1) Personalized Scene Generation, which places a given object in a scene that reflects a user's preferences and lifestyle, motivated by personalized product presentation; and (2) Personalized Creative Generation, which generates a novel image on a specified topic that is faithful to a user's aesthetic and visual identity, motivated by social media content creation. We further propose PEARL, which couples a multimodal reasoner with a frozen image generator in an interleaved reason-reflect loop optimized with differential data reward. Across both tasks, PEARL outperforms strong baselines, achieving an average improvement of 15% across personalization metrics.

Problem

• Personalized Image Generation
• User History
• Benchmark
• Personalization

Innovation

• Personalized Image Generation
• Reason-Reflect Loop
• Multimodal Reasoner
• Differential Data Reward
• User History Benchmark