HIDISC: A Hyperbolic Framework for Domain Generalization with Generalized Category Discovery

TL;DR

HIDISC introduces a hyperbolic learning framework for domain generalization and generalized category discovery, effectively handling distribution shifts and novel categories without relying on episodic training or target domain data.

Contribution

It proposes a novel hyperbolic representation method with GPT-guided augmentation and Tangent CutMix, enabling efficient domain and category generalization without synthetic domain simulation.

Findings

Achieves state-of-the-art results on PACS, Office-Home, and DomainNet datasets.

Outperforms existing Euclidean and hyperbolic GCD baselines.

Effectively handles distribution shifts and novel categories.

Abstract

Generalized Category Discovery (GCD) aims to classify test-time samples into either seen categories** -- available during training -- or novel ones, without relying on label supervision. Most existing GCD methods assume simultaneous access to labeled and unlabeled data during training and arising from the same domain, limiting applicability in open-world scenarios involving distribution shifts. Domain Generalization with GCD (DG-GCD) lifts this constraint by requiring models to generalize to unseen domains containing novel categories, without accessing targetdomain data during training. The only prior DG-GCD method, DG2CD-Net, relies on episodic training with multiple synthetic domains and task vector aggregation, incurring high computational cost and error accumulation. We propose HIDISC, a hyperbolic representation learning framework that achieves domain and category-level generalization without episodic simulation. To expose the model to minimal but diverse domain variations, we augment the source domain using GPT-guided diffusion, avoiding overfitting while maintaining efficiency. To structure the representation space, we introduce Tangent CutMix, a curvature-aware interpolation that synthesizes pseudo-novel samples in tangent space, preserving manifold consistency. A unified loss -- combining penalized Busemann alignment, hybrid hyperbolic contrastive regularization, and adaptive outlier repulsion -- **facilitates compact, semantically structured embeddings. A learnable curvature parameter further adapts the geometry to dataset complexity. HIDISC achieves state-of-the-art results on PACS , Office-Home , and DomainNet, consistently outperforming the existing Euclidean and hyperbolic (DG)-GCD baselines.