解决开放世界下类别发现中的捷径学习问题，提升模型泛化能力。
ClearGCD: Mitigating Shortcut Learning For Robust Generalized Category Discovery
- 通过跨类块替换生成强增强，结合弱增强保持语义一致性。
- 引入自适应原型库，同时对齐已知类并分离潜在新类。
- 可无缝集成现有方法，在多个基准上超越当前最优结果。
在开放世界场景中，广义类别发现（GCD）需要在无标签数据中识别已知和新类别。然而，现有方法常因捷径学习导致原型混淆，损害泛化性能并引发已知类遗忘。本文提出ClearGCD框架，通过两种互补机制缓解对非语义线索的依赖：首先，语义视图对齐（SVA）通过跨类块替换生成强增强，并利用弱增强强制语义一致性；其次，捷径抑制正则化（SSR）维护一个自适应原型库，既对齐已知类，又促进潜在新类的分离。ClearGCD可无缝集成至参数化GCD方法中，在多个基准上持续优于当前最优方法。
原文摘要 · Abstract (English)
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.
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。