Hypernym Detection Using Strict Partial Order Networks
This work addresses hypernym detection, a key task in natural language processing for tasks like taxonomy induction, with incremental improvements over existing methods.
The paper tackles the problem of detecting hypernymy relations by introducing Strict Partial Order Networks (SPON), a neural architecture that enforces asymmetry and transitivity as soft constraints, achieving state-of-the-art or superior performance on ten out of eleven benchmarks.
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.