TaxoBell: Gaussian Box Embeddings for Self-Supervised Taxonomy Expansion

TL;DR

TaxoBell introduces Gaussian box embeddings for automated taxonomy expansion, effectively modeling hierarchical and ambiguous relationships with improved stability and interpretability, outperforming existing methods on multiple benchmarks.

Contribution

It proposes a novel Gaussian box embedding framework that enhances stability, uncertainty modeling, and ambiguity handling in taxonomy expansion tasks.

Findings

Outperforms state-of-the-art baselines by 19% in MRR

Achieves around 25% improvement in Recall@k

Demonstrates robustness and interpretability through extensive experiments

Abstract

Taxonomies form the backbone of structured knowledge representation across diverse domains, enabling applications such as e-commerce catalogs, semantic search, and biomedical discovery. Yet, manual taxonomy expansion is labor-intensive and cannot keep pace with the emergence of new concepts. Existing automated methods rely on point-based vector embeddings, which model symmetric similarity and thus struggle with the asymmetric "is-a" relationships that are fundamental to taxonomies. Box embeddings offer a promising alternative by enabling containment and disjointness, but they face key issues: (i) unstable gradients at the intersection boundaries, (ii) no notion of semantic uncertainty, and (iii) limited capacity to represent polysemy or ambiguity. We address these shortcomings with TaxoBell, a Gaussian box embedding framework that translates between box geometries and multivariate Gaussian distributions, where means encode semantic location and covariances encode uncertainty. Energy-based optimization yields stable optimization, robust modeling of ambiguous concepts, and interpretable hierarchical reasoning. Extensive experimentation on five benchmark datasets demonstrates that TaxoBell significantly outperforms eight state-of-the-art taxonomy expansion baselines by 19% in MRR and around 25% in Recall@k. We further demonstrate the advantages and pitfalls of TaxoBell with error analysis and ablation studies.