ClearGCD: Mitigating Shortcut Learning For Robust Generalized Category Discovery
This addresses robust category discovery in open-world scenarios for computer vision applications, but it is incremental as it builds on existing parametric GCD approaches.
The paper tackles the problem of prototype confusion and shortcut learning in Generalized Category Discovery (GCD), which undermines generalization and causes forgetting of known classes, by proposing ClearGCD with Semantic View Alignment and Shortcut Suppression Regularization, resulting in consistent outperformance of state-of-the-art methods across multiple benchmarks.
In open-world scenarios, Generalized Category Discovery (GCD) requires identifying both known and novel categories within unlabeled data. However, existing methods often suffer from prototype confusion caused by shortcut learning, which undermines generalization and leads to forgetting of known classes. We propose ClearGCD, a framework designed to mitigate reliance on non-semantic cues through two complementary mechanisms. First, Semantic View Alignment (SVA) generates strong augmentations via cross-class patch replacement and enforces semantic consistency using weak augmentations. Second, Shortcut Suppression Regularization (SSR) maintains an adaptive prototype bank that aligns known classes while encouraging separation of potential novel ones. ClearGCD can be seamlessly integrated into parametric GCD approaches and consistently outperforms state-of-the-art methods across multiple benchmarks.