Is Everything in Order? A Simple Way to Order Sentences

TL;DR

This paper introduces Reorder-BART, a Transformer-based model that effectively orders shuffled sentences into coherent text, demonstrating state-of-the-art performance and strong generalization across multiple datasets.

Contribution

The paper presents Reorder-BART, a novel approach that formulates sentence ordering as a conditional text-to-marker generation task, achieving superior results over existing methods.

Findings

Re-BART outperforms previous models on 7 datasets in PMR and Kendall's tau.

The model generalizes well in zero-shot settings across different datasets.

Experiments reveal insights into the model's functioning and limitations.

Abstract

The task of organizing a shuffled set of sentences into a coherent text has been used to evaluate a machine's understanding of causal and temporal relations. We formulate the sentence ordering task as a conditional text-to-marker generation problem. We present Reorder-BART (Re-BART) that leverages a pre-trained Transformer-based model to identify a coherent order for a given set of shuffled sentences. The model takes a set of shuffled sentences with sentence-specific markers as input and generates a sequence of position markers of the sentences in the ordered text. Re-BART achieves the state-of-the-art performance across 7 datasets in Perfect Match Ratio (PMR) and Kendall's tau ($\tau$). We perform evaluations in a zero-shot setting, showcasing that our model is able to generalize well across other datasets. We additionally perform several experiments to understand the functioning and limitations of our framework.