Critical Patch-Aware Sparse Prompting with Decoupled Training for Continual Learning on the Edge
Published in IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026
Authors: Wonseon Lim, Jaesung Lee, Dae-Won Kim
We propose a critical patch-aware sparse prompting framework with decoupled training, designed to enable efficient continual learning under tight compute and memory budgets on edge devices.
Recommended citation: W. Lim, J. Lee, D. Kim. (2026). "Critical Patch-Aware Sparse Prompting with Decoupled Training for Continual Learning on the Edge." IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
Download Paper