Harnessing privileged information for hyperbole detection

Rhys BIDDLE, Maciej RYBIŃSKI, Qian LI, Cécile PARIS, Guandong XU

Research output: Chapter in Book/Report/Conference proceedingChapters

3 Citations (Scopus)

Abstract

The detection of hyperbole is an important stepping stone to understanding the intentions of a hyperbolic utterance. We propose a model that combines pre-trained language models with privileged information for the task of hyperbole detection. We also introduce a suite of behavioural tests to probe the capabilities of hyperbole detection models across a range of hyperbole types. Our experiments show that our model improves upon baseline models on an existing hyperbole detection dataset. Probing experiments combined with analysis using local linear approximations (LIME) show that our model excels at detecting one particular type of hyperbole. Further, we discover that our experiments highlight annotation artifacts introduced through the process of literal paraphrasing of hyperbole. These annotation artifacts are likely to be a roadblock to further improvements in hyperbole detection. Copyright © 2021 Australasian Language Technology Association.

Original languageEnglish
Title of host publicationProceedings of the The 19th Annual Workshop of the Australasian Language Technology Association
EditorsAfshin RAHIMI, William LANE, Guido ZUCCON
PublisherAustralasian Language Technology Association
Pages58-67
Volume19
Publication statusPublished - 2021

Citation

Biddle, R., Rybiński, M., Li, Q., Paris, C., & Xu, G. (2021). Harnessing privileged information for hyperbole detection. In A. Rahimi, W. Lane, & G. Zuccon (Eds.), Proceedings of the The 19th Annual Workshop of the Australasian Language Technology Association (pp. 58-67). Australasian Language Technology Association.

Fingerprint

Dive into the research topics of 'Harnessing privileged information for hyperbole detection'. Together they form a unique fingerprint.