HyperText: Endowing FastText with Hyperbolic Geometry
This work addresses the limitation of Euclidean space in capturing tree-like hierarchies for text classification, offering a more efficient model for NLP applications.
The authors tackled the problem of modeling hierarchical structures in natural language data by proposing HyperText, which integrates hyperbolic geometry into FastText, resulting in improved text classification performance with fewer parameters.
Natural language data exhibit tree-like hierarchical structures such as the hypernym-hyponym relations in WordNet. FastText, as the state-of-the-art text classifier based on shallow neural network in Euclidean space, may not model such hierarchies precisely with limited representation capacity. Considering that hyperbolic space is naturally suitable for modeling tree-like hierarchical data, we propose a new model named HyperText for efficient text classification by endowing FastText with hyperbolic geometry. Empirically, we show that HyperText outperforms FastText on a range of text classification tasks with much reduced parameters.