# Embedding time expressions for deep temporal ordering models

## TL;DR

This paper introduces a method to embed time expressions into neural models to improve temporal ordering of events, showing enhanced performance especially on datasets with frequent event-timex interactions.

## Contribution

It presents a novel framework for embedding timexes using a character LSTM trained on synthetic data, integrating explicit temporal signals into deep models.

## Key findings

- Small performance increase on MATRES dataset
- Substantial gains on dataset with frequent event-timex interactions
- Demonstrates the benefit of explicit temporal embeddings in neural models

## Abstract

Data-driven models have demonstrated state-of-the-art performance in inferring the temporal ordering of events in text. However, these models often overlook explicit temporal signals, such as dates and time windows. Rule-based methods can be used to identify the temporal links between these time expressions (timexes), but they fail to capture timexes' interactions with events and are hard to integrate with the distributed representations of neural net models. In this paper, we introduce a framework to infuse temporal awareness into such models by learning a pre-trained model to embed timexes. We generate synthetic data consisting of pairs of timexes, then train a character LSTM to learn embeddings and classify the timexes' temporal relation. We evaluate the utility of these embeddings in the context of a strong neural model for event temporal ordering, and show a small increase in performance on the MATRES dataset and more substantial gains on an automatically collected dataset with more frequent event-timex interactions.