RegD: Hierarchical Embeddings via Dissimilarity between Arbitrary Euclidean Regions

arXiv:2501.17518v3 Announce Type: replace-cross Abstract: Hierarchical data is common in many domains like life sciences and e-commerce, and its embeddings often play a critical role. While hyperbolic embeddings offer a theoretically grounded approach to representing hierarchies in low-dimensional spaces, current methods often rely on specific geometric constructs as embedding candidates. This reliance limits their generalizability and makes it difficult to integrate with techniques that model s