Deciphering Personalization: Towards Fine-Grained Explainability in Natural Language for Personalized Image Generation Models
This addresses the problem of opaque personalization in AI image generation for users, offering more precise explanations, though it is incremental as it builds on existing coarse-grained methods.
The paper tackles the lack of fine-grained explainability in personalized image generation models by introducing FineXL, which provides natural language descriptions and quantitative scores for multiple aspects of personalization, improving explainability accuracy by 56% in experiments.
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 \textbf{FineXL}, towards \textbf{Fine}-grained e\textbf{X}plainability in natural \textbf{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.