SubmissionNumber#=%=#39
FinalPaperTitle#=%=#WarwickNLP at SemEval-2024 Task 1: Low-Rank Cross-Encoders for Efficient Semantic Textual Relatedness
ShortPaperTitle#=%=#
NumberOfPages#=%=#7
CopyrightSigned#=%=#Fahad Ebrahim
JobTitle#==#
Organization#==#The University of Warwick, Coventry, UK
Abstract#==#This work participates in SemEval 2024 Task 1 on Semantic Textural Relatedness (STR) in Track A (supervised regression) in two languages, English and Moroccan Arabic. The task consists of providing a score of how two sentences relate to each other. The system developed in this work leveraged a cross-encoder with a merged fine-tuned Low-Rank Adapter (LoRA). The system was ranked eighth in English with a Spearman coefficient of 0.842, while Moroccan Arabic was ranked seventh with a score of 0.816. Moreover, various experiments were conducted to see the impact of different models and adapters on the performance and accuracy of the system.
Author{1}{Firstname}#=%=#Fahad
Author{1}{Lastname}#=%=#Ebrahim
Author{1}{Username}#=%=#bu_elfhood
Author{1}{Email}#=%=#fahad.ebrahim@warwick.ac.uk
Author{1}{Affiliation}#=%=#University of Warwick
Author{2}{Firstname}#=%=#Mike
Author{2}{Lastname}#=%=#Joy
Author{2}{Email}#=%=#M.S.Joy@warwick.ac.uk
Author{2}{Affiliation}#=%=#University of Warwick

==========
èéáğö