Enhancing Keyphrase Generation by BART Finetuning with Splitting and Shuffling

TL;DR

This paper introduces Keyphrase-Focused BART, a method that fine-tunes separate BART models for present and absent keyphrases, employing shuffling and ranking techniques to improve keyphrase generation performance.

Contribution

It presents a novel approach of separate BART fine-tuning for present and absent keyphrases, with effective shuffling and ranking strategies, achieving state-of-the-art results.

Findings

Achieved new state-of-the-art F1@5 scores on two benchmark datasets.

Effective separation of present and absent keyphrase modeling.

Shuffling and ranking methods improve keyphrase generation accuracy.

Abstract

Keyphrase generation is a task of identifying a set of phrases that best repre-sent the main topics or themes of a given text. Keyphrases are dividend int pre-sent and absent keyphrases. Recent approaches utilizing sequence-to-sequence models show effectiveness on absent keyphrase generation. However, the per-formance is still limited due to the hardness of finding absent keyphrases. In this paper, we propose Keyphrase-Focused BART, which exploits the differ-ences between present and absent keyphrase generations, and performs fine-tuning of two separate BART models for present and absent keyphrases. We further show effective approaches of shuffling keyphrases and candidate keyphrase ranking. For absent keyphrases, our Keyphrase-Focused BART achieved new state-of-the-art score on F1@5 in two out of five keyphrase gen-eration benchmark datasets.