Cross-Sample Relational Fusion: Unifying Domain Generalization and Class-Incremental Learning

TL;DR

This paper introduces CORF, a unified framework that enhances domain generalization and mitigates catastrophic forgetting in class-incremental learning by leveraging relational dependencies and adaptive sample weighting.

Contribution

CORF is a novel framework that combines selective sample refinement, confidence-based weighting, and cascaded distillation to improve continual learning across domains.

Findings

CORF achieves competitive results on benchmark datasets.

It effectively handles domain shifts and reduces forgetting.

The framework can be integrated with existing CIL algorithms.

Abstract

Class-Incremental Learning (CIL) requires a learning system to learn new classes while retaining previously learned knowledge. However, in real-world scenarios such as autonomous driving, a system trained on urban roads in sunny weather may later need to operate in rural or highway environments with different traffic patterns and weather conditions. This requires the model not only to overcome catastrophic forgetting, but also to effectively handle domain shifts. In this paper, we propose CrOss-sample Relational Fusion (CORF), a unified framework to address domain shift and catastrophic forgetting simultaneously. To enhance generalizability, we perform selective refinement of training samples by leveraging spatial contribution maps to highlight semantically informative regions. Furthermore, we incorporate predictive confidence to adaptively weigh samples, thereby facilitating the learning of domain-agnostic representations. To alleviate forgetting, we propose a cascaded distillation framework that captures cross-sample relational dependencies across multiple feature hierarchies, enabling multi-grained knowledge transfer from previous tasks. CORF can be seamlessly integrated into existing CIL algorithms to enhance their generalizability, achieving competitive performance across various benchmark datasets. Code is available at https://github.com/LAMDA-CL/TMM26-CORF .