arXiv:2512.09773v1 Announce Type: new 
Abstract: We introduce Stylized Meta-Album (SMA), a new image classification meta-dataset comprising 24 datasets (12 content datasets, and 12 stylized datasets), designed to advance studies on out-of-distribution (OOD) generalization and related topics. Created using style transfer techniques from 12 subject classification datasets, SMA provides a diverse and extensive set of 4800 groups, combining various subjects (objects, plants, animals, human actions, textures) with multiple styles. SMA enables flexible control over groups and classes, allowing us to configure datasets to reflect diverse benchmark scenarios. While ideally, data collection would capture extensive group diversity, practical constraints often make this infeasible. SMA addresses this by enabling large and configurable group structures through flexible control over styles, subject classes, and domains-allowing datasets to reflect a wide range of real-world benchmark scenarios. This design not only expands group and class diversity, but also opens new methodological directions for evaluating model performance across diverse group and domain configurations-including scenarios with many minority groups, varying group imbalance, and complex domain shifts-and for studying fairness, robustness, and adaptation under a broader range of realistic conditions. To demonstrate SMA's effectiveness, we implemented two benchmarks: (1) a novel OOD generalization and group fairness benchmark leveraging SMA's domain, class, and group diversity to evaluate existing benchmarks. Our findings reveal that while simple balancing and algorithms utilizing group information remain competitive as claimed in previous benchmarks, increasing group diversity significantly impacts fairness, altering the superiority and relative rankings of algorithms. We also propose to use \textit{Top-M worst group accuracy} as a new hyperparameter tuning metric, demonstrating broader fairness during optimization and delivering better final worst-group accuracy for larger group diversity. (2) An unsupervised domain adaptation (UDA) benchmark utilizing SMA's group diversity to evaluate UDA algorithms across more scenarios, offering a more comprehensive benchmark with lower error bars (reduced by 73\% and 28\% in closed-set setting and UniDA setting, respectively) compared to existing efforts. These use cases highlight SMA's potential to significantly impact the outcomes of conventional benchmarks.

تم تقديم الألبوم المصور (SMA) كبيانات جديدة لتصنيف الصور، ويتكون من 24 مجموعة بيانات مصممة لتعزيز الدراسات حول التعميم خارج التوزيع (OOD). يتضمن هذا المجموعة 4800 مجموعة تم إنشاؤها من خلال تقنيات نقل الأسلوب، مما يسمح بالتحكم المرن في مواضيع وأنماط متعددة.

Se ha presentado el Stylized Meta-Album (SMA) como un nuevo meta-conjunto de datos de clasificación de imágenes, que comprende 24 conjuntos de datos diseñados para mejorar los estudios sobre la generalización fuera de distribución (OOD). Este conjunto de datos incluye 4800 grupos creados mediante técnicas de transferencia de estilo, permitiendo un control flexible sobre varios sujetos y estilos.

Le Stylized Meta-Album (SMA) a été introduit comme un nouveau méta-ensemble de données de classification d'images, comprenant 24 ensembles de données conçus pour améliorer les études sur la généralisation hors distribution (OOD). Cet ensemble de données inclut 4800 groupes créés grâce à des techniques de transfert de style, permettant un contrôle flexible sur divers sujets et styles.

The Stylized Meta-Album (SMA) has been introduced as a new image classification meta-dataset, comprising 24 datasets designed to enhance studies on out-of-distribution (OOD) generalization. This dataset includes 4800 groups created through style transfer techniques, allowing for flexible control over various subjects and styles.

Stylized Meta-Album: Group-bias injection with style transfer to study robustness against distribution shifts

One More Thing in AI – Your Shortcut to AI Mastery

Stylized Meta-Album: Group-bias injection with style transfer to study robustness against distribution shifts

Was this article worth reading? Share it

One More Thing in AI

SwapAnything.io

Kosmik

Stockfill

Srefs

Blunge

Ready to build your own newsroom?