Distilling Hypernymy Relations from Language Models: On the Effectiveness of Zero-Shot Taxonomy Induction
This work addresses the problem of automating taxonomy induction for NLP applications, offering an incremental improvement by leveraging language models in a zero-shot setting.
The paper tackled zero-shot taxonomy induction by distilling hypernymy relations from language models via prompting and sentence scoring, showing that these simple methods outperform some supervised strategies and are competitive with state-of-the-art under certain conditions.
In this paper, we analyze zero-shot taxonomy learning methods which are based on distilling knowledge from language models via prompting and sentence scoring. We show that, despite their simplicity, these methods outperform some supervised strategies and are competitive with the current state-of-the-art under adequate conditions. We also show that statistical and linguistic properties of prompts dictate downstream performance.