Hypernym Detection Using Strict Partial Order Networks

TL;DR

This paper presents SPON, a neural network architecture that enforces logical properties for hypernym detection, achieving state-of-the-art results across multiple benchmarks.

Contribution

Introduction of SPON, a neural network enforcing asymmetry and transitivity for hypernym detection, with an augmented version for generalizing to unseen terms.

Findings

SPON outperforms existing methods on most benchmarks.

The augmented SPON generalizes well to unseen terms.

Consistent improvement across eleven diverse datasets.

Abstract

This paper introduces Strict Partial Order Networks (SPON), a novel neural network architecture designed to enforce asymmetry and transitive properties as soft constraints. We apply it to induce hypernymy relations by training with is-a pairs. We also present an augmented variant of SPON that can generalize type information learned for in-vocabulary terms to previously unseen ones. An extensive evaluation over eleven benchmarks across different tasks shows that SPON consistently either outperforms or attains the state of the art on all but one of these benchmarks.