Semantic Pivoting Model for Effective Event Detection

TL;DR

This paper introduces SPEED, a semantic pivoting model that enhances event detection by leveraging semantic correlations, achieving state-of-the-art results without external resources.

Contribution

The paper proposes a novel semantic pivoting approach that explicitly incorporates prior semantic information for improved event detection.

Findings

Achieves state-of-the-art performance on event detection tasks.

Outperforms baseline models in multiple experimental settings.

Does not require external resources for training.

Abstract

Event Detection, which aims to identify and classify mentions of event instances from unstructured articles, is an important task in Natural Language Processing (NLP). Existing techniques for event detection only use homogeneous one-hot vectors to represent the event type classes, ignoring the fact that the semantic meaning of the types is important to the task. Such an approach is inefficient and prone to overfitting. In this paper, we propose a Semantic Pivoting Model for Effective Event Detection (SPEED), which explicitly incorporates prior information during training and captures semantically meaningful correlations between input and events. Experimental results show that our proposed model achieves state-of-the-art performance and outperforms the baselines in multiple settings without using any external resources.