MIDAS at SemEval-2020 Task 10: Emphasis Selection using Label Distribution Learning and Contextual Embeddings

TL;DR

This paper introduces a method for emphasis selection in text using contextual embeddings and label distribution learning, achieving competitive results and analyzing linguistic factors affecting performance.

Contribution

It presents a novel approach combining contextual embeddings with label distribution learning for emphasis selection, and explores ensemble models and linguistic analysis.

Findings

Best ensemble model achieved a 0.783 score

Method effectively handles annotator disagreements

Analysis reveals influence of POS tags and sentence length

Abstract

This paper presents our submission to the SemEval 2020 - Task 10 on emphasis selection in written text. We approach this emphasis selection problem as a sequence labeling task where we represent the underlying text with various contextual embedding models. We also employ label distribution learning to account for annotator disagreements. We experiment with the choice of model architectures, trainability of layers, and different contextual embeddings. Our best performing architecture is an ensemble of different models, which achieved an overall matching score of 0.783, placing us 15th out of 31 participating teams. Lastly, we analyze the results in terms of parts of speech tags, sentence lengths, and word ordering.