Personalized Image Aesthetic Assessment via Preference-rich Sample Mining and Cohort Merging

Zhichao Yang1†, Tianjiao Gu1†, Zhixianhe Zhang1, Xiangfei Sheng1, Pengfei Chen1, Leida Li1,2*
1School of Artificial Intelligence, Xidian University   2State Key Laboratory of EMIM, Xidian University
†Equal contribution   *Corresponding author
arXiv Code

🤗

 Model
介绍图

Abstract

Personalized Image Aesthetic Assessment (PIAA) aims to predict aesthetic ratings of images that vary across individuals. The aesthetic preferences manifest to different extents across distinct visual stimuli and exhibit cohort-specific patterns. Motivated by the above fact, this paper presents a Multimodal Large Language Model (MLLM)-based approach, which models individual aesthetic preferences by Preference-Rich sample mining and Aesthetically-resonant Cohort merging (PRAC). Specifically, PRAC first identifies preference-rich samples by analyzing both Collective Controversy and Personalized Deviation of images, maximizing the utility of limited user data. Based upon the preference-rich samples, cross-user preference similarities are measured by comparing preference embeddings. Then, a cohort-based model merging strategy, is proposed by aggregating preference patterns from aesthetically-resonant users, which further enhances the personalization for the target individual. Extensive experiments and comparisons on four benchmark PIAA databases demonstrate the superiority of the proposed PRAC model over the state-of-the-arts.

Motivation

Insight-1: Psychological research has demonstrated that some stimuli naturally elicit more active preference responses than others, indicating that not all images are equally informative for modeling aesthetic taste.
Insight-2: Researches in cognitive aesthetics revealed that aesthetic preferences often cluster among individuals with shared sensibilities—beauty is in the eye of your cohort—suggesting that identifying aesthetically-resonant cohort offers valuable insights for target user modeling.

algebraic reasoning

PRAC Model

Overview of the proposed PRAC. The framework consists of three stages:
(a) Generic Aesthetic Predictor: the MLLM is trained with aesthetic distribution supervision on generic aesthetic data to establish foundational aesthetic understanding.
(b) PreferSelect: preference-rich samples are identified through collective controversy and personalized deviation metrics.
(c) PreferMerge: the target user’s personalized model is constructed by merging models from aesthetically-resonant cohort.

grade-lv

Quantitative Results

Performance comparison between the proposed PRAC and state-of-the-art PIAA methods on PARA, FlickrAES and AADB.

grade-lv

Qualitative Results

A qualitative study using two example users to visualize how the proposed PRAC model performs sample mining and cohort merging, ultimately facilitating personalized aesthetic expression.

grade-lv

BibTeX


@article{yang2026personalized,
  title={Personalized Image Aesthetic Assessment via Preference-rich Sample Mining and Cohort Merging},
  author={Yang, Zhichao and Gu, tianjiao and Zhang, Zhixianhe and Sheng, Xiangfei and Chen, Pengfei and Li, Leida},
  booktitle={Proceedings of the 34th ACM International Conference on Multimedia},
  year={2026}
}