Search papers, labs, and topics across Lattice.
The paper addresses Continual Generalized Category Discovery (C-GCD) by proposing GOAL, a framework with a fixed Equiangular Tight Frame (ETF) classifier to maintain consistent geometric structure during learning. GOAL performs supervised alignment for labeled data and confidence-guided alignment for unlabeled data, mitigating forgetting and enabling stable incorporation of new classes. Experiments on four benchmarks demonstrate that GOAL significantly outperforms the prior state-of-the-art method, Happy, improving both knowledge retention and novel class discovery.
Achieve stable continual learning without catastrophic forgetting by fixing classifier weights to an Equiangular Tight Frame and aligning features geometrically.
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.
