通过测试阶段语义演化，恢复增量学习中遗忘的知识。
Restoring Forgotten Knowledge in Non-Exemplar Class Incremental Learning through Test-Time Semantic Evolution
- 在测试时通过自监督方式动态修正语义漂移
- 在多个数据集上超越主流方法，冷启动与热启动均有效
- 无需梯度下降，提供解析解提升在线测试效率
持续学习旨在随数据流积累知识并缓解灾难性遗忘。在非样本类增量学习（NECIL）中，由于旧类别不可访问，增量优化过程导致知识遗忘。现有方法难以在训练阶段平衡稳定性与可塑性。我们发现测试阶段被忽视，却可能是解决遗忘的突破口。为此提出RoSE：一种基于测试时语义演化的知识恢复方法。该方法设计用于最小化遗忘，是一种自监督的语义漂移补偿框架，能更准确估计漂移。为避免在线测试中优化不完整，我们推导出无需梯度下降的解析解。在CIFAR-100、TinyImageNet和ImageNet100上评估，涵盖冷启动与热启动场景，结果表明本方法在各类场景下均持续优于多数最先进方法，验证了测试时演化在NECIL中的潜力与可行性。
原文摘要 · Abstract (English)
Continual learning aims to accumulate knowledge over a data stream while mitigating catastrophic forgetting. In Non-exemplar Class Incremental Learning (NECIL), forgetting arises during incremental optimization because old classes are inaccessible, hindering the retention of prior knowledge. To solve this, previous methods struggle in achieving the stability-plasticity balance in the training stages. However, we note that the testing stage is rarely considered among them, but is promising to be a solution to forgetting. Therefore, we propose RoSE, which is a simple yet effective method that \textbf{R}est\textbf{o}res forgotten knowledge through test-time \textbf{S}emantic \textbf{E}volution. Specifically designed for minimizing forgetting, RoSE is a test-time semantic drift compensation framework that enables more accurate drift estimation in a self-supervised manner. Moreover, to avoid incomplete optimization during online testing, we derive an analytical solution as an alternative to gradient descent. We evaluate RoSE on CIFAR-100, TinyImageNet, and ImageNet100 datasets, under both cold-start and warm-start settings. Our method consistently outperforms most state-of-the-art (SOTA) methods across various scenarios, validating the potential and feasibility of test-time evolution in NECIL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。