Reproduce “Task-Adaptive Saliency Guidance for Exemplar-free Class Incremental Learning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Xialei Liu, Jiang-Tian Zhai, Andrew D. Bagdanov, Ke Li, Ming-Ming Cheng
The authors introduce TASS for exemplar-free class-incremental learning, guiding where a classifier attends without retaining earlier-task images. It combines low-level saliency supervision, task-adaptive boundary guidance, and injected saliency-like noise. On CIFAR-100 with 10 tasks, TASS integrated into SSRE reaches 57.93% last top-1 accuracy, versus 55.01% for SSRE; this benchmark result does not prove saliency drift alone causes forgetting.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.