TaxoBell: Gaussian Box Embeddings for Self-Supervised Taxonomy Expansion
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916062863294464 |
|---|---|
| author | Mishra, Sahil Srinivasan, Srinitish Bedathur, Srikanta Chakraborty, Tanmoy |
| author_facet | Mishra, Sahil Srinivasan, Srinitish Bedathur, Srikanta Chakraborty, Tanmoy |
| contents | Taxonomies form the backbone of structured knowledge representation across diverse domains, enabling applications such as e-commerce and semantic search. Yet, manual taxonomy expansion is labor-intensive and slow. Existing methods rely on point-based vector embeddings, which model symmetric similarity and thus struggle with the asymmetric 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 experiments on five benchmark datasets demonstrate 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_09633 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | TaxoBell: Gaussian Box Embeddings for Self-Supervised Taxonomy Expansion Mishra, Sahil Srinivasan, Srinitish Bedathur, Srikanta Chakraborty, Tanmoy Computation and Language Taxonomies form the backbone of structured knowledge representation across diverse domains, enabling applications such as e-commerce and semantic search. Yet, manual taxonomy expansion is labor-intensive and slow. Existing methods rely on point-based vector embeddings, which model symmetric similarity and thus struggle with the asymmetric 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 experiments on five benchmark datasets demonstrate 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. |
| title | TaxoBell: Gaussian Box Embeddings for Self-Supervised Taxonomy Expansion |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.09633 |