Deciphering Personalization: Towards Fine-Grained Explainability in Natural Language for Personalized Image Generation Models

2025-11-02

AI Summary

Current personalized image generation models lack fine-grained, interpretable personalization mechanisms; natural language explanations are typically coarse-grained and insufficient for characterizing multi-dimensional semantic attributes and their intensity variations. To address this, we propose FineXL—the first method enabling joint identification and quantitative interpretation of multiple semantic dimensions (e.g., style, composition, texture) and their intensities in personalized generation. FineXL integrates vision-language alignment, feature disentanglement analysis, and hierarchical attribution to extract salient semantic factors and generate both natural language descriptions and numerical intensity scores. Extensive experiments across diverse generative models and application scenarios demonstrate that FineXL improves explanation accuracy by 56% over prior approaches, significantly enhancing user understanding of and control over the personalization decision process.

Technology Category

• Natural Language Processing: Generation
• Humans and AI: Explainable AI (XAI) for Human Understanding
• Computer Vision: Large Vision Models

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Image generation models are usually personalized in practical uses in order to better meet the individual users'heterogeneous needs, but most personalized models lack explainability about how they are being personalized. Such explainability can be provided via visual features in generated images, but is difficult for human users to understand. Explainability in natural language is a better choice, but the existing approaches to explainability in natural language are limited to be coarse-grained. They are unable to precisely identify the multiple aspects of personalization, as well as the varying levels of personalization in each aspect. To address such limitation, in this paper we present a new technique, namely extbf{FineXL}, towards extbf{Fine}-grained e extbf{X}plainability in natural extbf{L}anguage for personalized image generation models. FineXL can provide natural language descriptions about each distinct aspect of personalization, along with quantitative scores indicating the level of each aspect of personalization. Experiment results show that FineXL can improve the accuracy of explainability by 56%, when different personalization scenarios are applied to multiple types of image generation models.

Problem

• Providing fine-grained natural language explanations for personalized image generation models
• Identifying multiple distinct aspects and levels of personalization in generated images
• Improving explainability accuracy for heterogeneous user needs in AI systems

Innovation

• Fine-grained explainability in natural language
• Quantitative personalization scores per aspect
• Improved accuracy by 56% in experiments