Dynamically Updating Event Representations for Temporal Relation Classification with Multi-category Learning

TL;DR

This paper introduces a dynamic, event-centric model for temporal relation classification that leverages multi-task learning to improve performance across multiple TLINK categories in English and Japanese datasets.

Contribution

The paper proposes a novel event-centric, multi-task learning model that dynamically updates event representations, addressing limitations of independent classifiers in temporal relation classification.

Findings

Outperforms state-of-the-art models on English and Japanese data.

Effectively leverages full dataset size through multi-task learning.

Demonstrates improved accuracy in temporal relation classification.

Abstract

Temporal relation classification is a pair-wise task for identifying the relation of a temporal link (TLINK) between two mentions, i.e. event, time, and document creation time (DCT). It leads to two crucial limits: 1) Two TLINKs involving a common mention do not share information. 2) Existing models with independent classifiers for each TLINK category (E2E, E2T, and E2D) hinder from using the whole data. This paper presents an event centric model that allows to manage dynamic event representations across multiple TLINKs. Our model deals with three TLINK categories with multi-task learning to leverage the full size of data. The experimental results show that our proposal outperforms state-of-the-art models and two transfer learning baselines on both the English and Japanese data.