SubmissionNumber#=%=#67
FinalPaperTitle#=%=#USMBA-NLP at SemEval-2024 Task 2: Safe Biomedical Natural Language Inference for Clinical Trials using Bert
ShortPaperTitle#=%=#
NumberOfPages#=%=#5
CopyrightSigned#=%=#Anass Fahfouh
JobTitle#==#
Organization#==#
Abstract#==#This paper presents the application of BERT in
SemEval 2024 Task 2, Safe Biomedical Natu-
ral Language Inference for Clinical Trials. The
main objectives of this task were: First, to in-
vestigate the consistency of BERT in its rep-
resentation of semantic phenomena necessary
for complex inference in clinical NLI settings.
Second, to investigate the ability of BERT to
perform faithful reasoning, i.e., make correct
predictions for the correct reasons. The submit-
ted model is fine-tuned on the NLI4CT dataset,
which is enhanced with a novel contrast set,
using binary cross entropy loss.
Author{1}{Firstname}#=%=#Anass
Author{1}{Lastname}#=%=#Fahfouh
Author{1}{Username}#=%=#anass
Author{1}{Email}#=%=#anass.fh10@gmail.com
Author{1}{Affiliation}#=%=#FSDM
Author{2}{Firstname}#=%=#Abdessamad
Author{2}{Lastname}#=%=#Benlahbib
Author{2}{Email}#=%=#abdessamad.benlahbib@usmba.ac.ma
Author{2}{Affiliation}#=%=#LISAC
Author{3}{Firstname}#=%=#Jamal
Author{3}{Lastname}#=%=#Riffi
Author{3}{Email}#=%=#riffi.jamal@gmail.com
Author{3}{Affiliation}#=%=#LISAC
Author{4}{Firstname}#=%=#Hamid
Author{4}{Lastname}#=%=#Tairi
Author{4}{Email}#=%=#htairi@yahoo.fr
Author{4}{Affiliation}#=%=#LISAC

==========
èéáğö