Continual Generalized Category Discovery (C-GCD) requires identifying novel classes from unlabeled data while retaining knowledge of known classes over time. Existing methods typically update classifier weights dynamically, resulting in forgetting and inconsistent feature alignment. We propose GOAL, a unified framework that introduces a fixed Equiangular Tight Frame (ETF) classifier to impose a consistent geometric structure throughout learning. GOAL conducts supervised alignment for labeled samples and confidence-guided alignment for novel samples, enabling stable integration of new classes without disrupting old ones. Experiments on four benchmarks show that GOAL outperforms the prior method Happy, reducing forgetting by 16.1% and boosting novel class discovery by 3.2%, establishing a strong solution for long-horizon continual discovery.

翻译：持续广义类别发现（C-GCD）要求在随时间推移保留已知类别知识的同时，从无标注数据中识别新类别。现有方法通常动态更新分类器权重，导致遗忘和特征对齐不一致。我们提出GOAL，一个引入固定等角紧框架（ETF）分类器的统一框架，以在整个学习过程中施加一致的几何结构。GOAL对标注样本进行监督对齐，对新样本进行置信度引导的对齐，从而在不破坏旧类别的情况下稳定地整合新类别。在四个基准测试上的实验表明，GOAL优于现有方法Happy，将遗忘率降低了16.1%，并将新类别发现率提高了3.2%，为长时程持续发现提供了一个强有力的解决方案。