# Recognizing Arrow Of Time In The Short Stories

## TL;DR

This paper introduces a new dataset and demonstrates that pre-trained BERT models can effectively recognize the chronological order in short stories, outperforming RNN-based methods.

## Contribution

The paper presents a novel dataset for arrow of time recognition in short stories and evaluates BERT's effectiveness over RNN architectures.

## Key findings

- BERT achieves reasonable accuracy on the task.
- BERT outperforms RNN-based architectures.
- The dataset facilitates research on temporal understanding in narratives.

## Abstract

Recognizing arrow of time in short stories is a challenging task. i.e., given only two paragraphs, determining which comes first and which comes next is a difficult task even for humans. In this paper, we have collected and curated a novel dataset for tackling this challenging task. We have shown that a pre-trained BERT architecture achieves reasonable accuracy on the task, and outperforms RNN-based architectures.