OntoType: Ontology-Guided and Pre-Trained Language Model Assisted Fine-Grained Entity Typing
Tanay Komarlu, Minhao Jiang, Xuan Wang, Jiawei Han
TL;DR
OntoType tackles the data bottleneck in fine-grained entity typing by offering an annotation-free, ontology-guided framework that leverages multiple PLM prompts and an NLI-based refinement strategy. It first generates candidate types from head words and MLM prompts, then aligns them to a coarse ontology level via semantic similarity, and finally descends the type hierarchy using a ranking score defined by $rank(type) = \sigma_{entail} + \sigma_{cand}$ to obtain fine-grained labels. Experiments on Ontonotes, FIGER, and NYT demonstrate strong zero-shot performance, with notable gains on NYT (5.9 Ma-F1 over ZOE) and substantial improvements over prior zero-shot methods, while remaining competitive with supervised baselines on some datasets. The results underscore the value of a well-structured ontological hierarchy for disambiguation and fine-grained typing, and point to ontology refinement and nested-entity handling as meaningful future directions.
Abstract
Fine-grained entity typing (FET), which assigns entities in text with context-sensitive, fine-grained semantic types, is a basic but important task for knowledge extraction from unstructured text. FET has been studied extensively in natural language processing and typically relies on human-annotated corpora for training, which is costly and difficult to scale. Recent studies explore the utilization of pre-trained language models (PLMs) as a knowledge base to generate rich and context-aware weak supervision for FET. However, a PLM still requires direction and guidance to serve as a knowledge base as they often generate a mixture of rough and fine-grained types, or tokens unsuitable for typing. In this study, we vision that an ontology provides a semantics-rich, hierarchical structure, which will help select the best results generated by multiple PLM models and head words. Specifically, we propose a novel annotation-free, ontology-guided FET method, OntoType, which follows a type ontological structure, from coarse to fine, ensembles multiple PLM prompting results to generate a set of type candidates, and refines its type resolution, under the local context with a natural language inference model. Our experiments on the Ontonotes, FIGER, and NYT datasets using their associated ontological structures demonstrate that our method outperforms the state-of-the-art zero-shot fine-grained entity typing methods as well as a typical LLM method, ChatGPT. Our error analysis shows that refinement of the existing ontology structures will further improve fine-grained entity typing.