SubmissionNumber#=%=#168
FinalPaperTitle#=%=#IUST-NLPLAB at SemEval-2024 Task 9: BRAINTEASER By MPNet (Sentence Puzzle)
ShortPaperTitle#=%=#
NumberOfPages#=%=#4
CopyrightSigned#=%=#Mohammad Hossein Abbaspour
JobTitle#==#
Organization#==#Iran University of Science and Technology
Abstract#==#This study addresses a task encompassing two distinct subtasks: Sentence-puzzle and Word-puzzle. Our primary focus lies within the Sentence-puzzle subtask, which involves discerning the correct answer from a set of three options for a given riddle constructed from sentence fragments. We propose four distinct methodologies tailored to address this subtask effectively. Firstly, we introduce a zero-shot approach leveraging the capabilities of the GPT-3.5 model. Additionally, we present three fine-tuning methodologies utilizing MPNet as the underlying architecture, each employing a different loss function. We conduct comprehensive evaluations of these methodologies on the designated task dataset and meticulously document the obtained results. Furthermore, we conduct an in-depth analysis to ascertain the respective strengths and weaknesses of each method. Through this analysis, we aim to provide valuable insights into the challenges inherent to this task domain.
Author{1}{Firstname}#=%=#Mohammad Hossein
Author{1}{Lastname}#=%=#Abbaspour
Author{1}{Username}#=%=#mohammadhabp
Author{1}{Email}#=%=#mohammadhabp@gmail.com
Author{1}{Affiliation}#=%=#Iran University of Science and Technology
Author{2}{Firstname}#=%=#Erfan
Author{2}{Lastname}#=%=#Moosavi Monazzah
Author{2}{Username}#=%=#erfanmoosavimonazzah
Author{2}{Email}#=%=#erfan.moosavi.monazzah@gmail.com
Author{2}{Affiliation}#=%=#Iran University of Science and Technology
Author{3}{Firstname}#=%=#Sauleh
Author{3}{Lastname}#=%=#Eetemadi
Author{3}{Username}#=%=#sauleh
Author{3}{Email}#=%=#sauleh@iust.ac.ir
Author{3}{Affiliation}#=%=#Iran University of Science and Technology

==========
èéáğö