Maintaining Consistent Inter-Class Topology in Continual Test-Time Adaptation

2025-01-01 · CVPR 2025 1 · Chenggong Ni, Fan Lyu, Jiayao Tan, Fuyuan Hu, Rui Yao, Tao Zhou

This paper introduces Topological Consistency Adaptation (TCA), a novel approach to Continual Test-time Adaptation (CTTA) that addresses the challenges of domain shifts and error accumulation in testing scenarios. TCA ensures the stability of inter-class relationships by enforcing a class topological consistency constraint, which minimizes the distortion of class centroids and preserves the topological structure during continuous adaptation. Additionally, we propose an intra-class compactness loss to maintain compactness within classes, indirectly supporting inter-class stability. To further enhance model adaptation, we introduce a batch imbalance topology weighting mechanism that accounts for class distribution imbalances within each batch, optimizing centroid distances and stabilizing the inter-class topology. Experiments show that our method demonstrates improvements in handling continuous domain shifts, ensuring stable feature distributions and boosting predictive performance.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Test-time Adaptation

MasFACT: Continual Multi-Agent Topology Learning via Geometry-Aware Posterior Transfer

2026-05-17 · Xuefei Wang, Jialu Wang, Fengbo Zhang, Yihan Hu 외 arxiv

Multi-agent systems (MAS) powered by large language models (LLMs) have emerged as a powerful paradigm for complex problem solving, where performance critically depends on the underlying inter-agent communication topology…

Inductive Graph Few-shot Class Incremental Learning

2024-11-11 · Yayong Li, Peyman Moghadam, Can Peng, Nan Ye 외

Node classification with Graph Neural Networks (GNN) under a fixed set of labels is well known in contrast to Graph Few-Shot Class Incremental Learning (GFSCIL), which involves learning a GNN classifier as graph nodes an…

class-incremental learning Class Incremental Learning Few-Shot Class-Incremental Learning Incremental Learning+1

CTP: Towards Vision-Language Continual Pretraining via Compatible Momentum Contrast and Topology Preservation

2023-08-14 · Hongguang Zhu, Yunchao Wei, Xiaodan Liang, Chunjie Zhang 외

Vision-Language Pretraining (VLP) has shown impressive results on diverse downstream tasks by offline training on large-scale datasets. Regarding the growing nature of real-world data, such an offline training paradigm o…

Continual Learning Continual Pretraining

CTP:Towards Vision-Language Continual Pretraining via Compatible Momentum Contrast and Topology Preservation

2023-01-01 · ICCV 2023 1 · Hongguang Zhu, Yunchao Wei, Xiaodan Liang, Chunjie Zhang 외

Continual Learning Continual Pretraining

Distributed Continual Learning

2024-05-23 · Long Le, Marcel Hussing, Eric Eaton

This work studies the intersection of continual and federated learning, in which independent agents face unique tasks in their environments and incrementally develop and share knowledge. We introduce a mathematical frame…

Continual Learning Federated Learning