CogNLP-Sheffield at CMCL 2021 Shared Task: Blending cognitively inspired features with transformer-based language models for predicting eye tracking patterns

Vickers, P., Wainwright, R., Tayyar Madabushi, H. et al. (1 more author) (2021) CogNLP-Sheffield at CMCL 2021 Shared Task: Blending cognitively inspired features with transformer-based language models for predicting eye tracking patterns. In: Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics (CMCL 2021). Cognitive Modeling and Computational Linguistics (CMCL 2021), 10 Jun 2021, Mexico City, Mexico. Association for Computational Linguistics (ACL), pp. 125-133. ISBN: 9781954085350.

Abstract

The CogNLP-Sheffield submissions to the CMCL 2021 Shared Task examine the value of a variety of cognitively and linguistically inspired features for predicting eye tracking patterns, as both standalone model inputs and as supplements to contextual word embeddings (XLNet). Surprisingly, the smaller pre-trained model (XLNet-base) outperforms the larger (XLNet-large), and despite evidence that multi-word expressions (MWEs) provide cognitive processing advantages, MWE features provide little benefit to either model.

Metadata

Item Type:	Proceedings Paper
Authors/Creators:	Vickers, P. Wainwright, R. Tayyar Madabushi, H. Villavicencio, A. https://orcid.org/0000-0002-3731-9168
Copyright, Publisher and Additional Information:	© 2021 The Authors. Licensed on a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).
Dates:	Accepted: 30 April 2021 Published (online): June 2021 Published: June 2021
Institution:	The University of Sheffield
Academic Units:	The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Computer Science (Sheffield)
Funding Information:	Funder Grant number Engineering and Physical Science Research Council EP/T02450X/1; 2269013 The Royal Society NAF\R2\202209
Date Deposited:	07 May 2021 11:51
Last Modified:	01 Feb 2022 18:08
Status:	Published
Publisher:	Association for Computational Linguistics (ACL)
Refereed:	Yes
Identification Number:	10.18653/v1/2021.cmcl-1.16
Related URLs:	Conference
Open Archives Initiative ID (OAI ID):	oai:eprints.whiterose.ac.uk:173802

Download

Published Version

Filename: 2021.cmcl-1.16.pdf

Licence: CC-BY 4.0

CLICK TO DOWNLOAD

CORE (COnnecting REpositories)

CogNLP-Sheffield at CMCL 2021 Shared Task: Blending cognitively inspired features with transformer-based language models for predicting eye tracking patterns

Abstract

Metadata

Download

Published Version

Export

Statistics