TA-DA: Topic-Aware Domain Adaptation for Scientific Keyphrase Identification and Classification (Student Abstract)

TL;DR

TA-DA is a novel framework that enhances scientific keyphrase extraction by combining topic-awareness, domain adaptation, multi-task learning, and adversarial training, leading to improved performance over baseline models.

Contribution

It introduces TA-DA, a new domain adaptation framework that integrates multiple techniques for better scientific keyphrase extraction.

Findings

Up to 5% improvement in F1-score over baselines

Effective integration of multi-task learning and adversarial training

Enhanced performance on scientific document keyphrase extraction

Abstract

Keyphrase identification and classification is a Natural Language Processing and Information Retrieval task that involves extracting relevant groups of words from a given text related to the main topic. In this work, we focus on extracting keyphrases from scientific documents. We introduce TA-DA, a Topic-Aware Domain Adaptation framework for keyphrase extraction that integrates Multi-Task Learning with Adversarial Training and Domain Adaptation. Our approach improves performance over baseline models by up to 5% in the exact match of the F1-score.