SubmissionNumber#=%=#261
FinalPaperTitle#=%=#Team MGTD4ADL at SemEval-2024 Task 8: Leveraging (Sentence) Transformer Models with Contrastive Learning for Identifying Machine-Generated Text
ShortPaperTitle#=%=#
NumberOfPages#=%=#8
CopyrightSigned#=%=#Ercong Nie
JobTitle#==#
Organization#==#Center for Information and Language Processing, Ludwig Maximilians University of Munich
Oettingenstraße 67, 80538 Munich, Germany
Abstract#==#This paper outlines our approach to SemEval-2024 Task 8 (Subtask B), which focuses on discerning machine-generated text from human-written content, while also identifying the text sources, i.e., from which Large Language Model (LLM) the target text is generated. Our detection system is built upon Transformer-based techniques, leveraging various pre-trained language models (PLMs), including sentence transformer models. Additionally, we incorporate Contrastive Learning (CL) into the classifier to improve the detecting capabilities and employ Data Augmentation methods. Ultimately, our system achieves a peak accuracy of 76.96% on the test set of the competition, configured using a sentence transformer model integrated with CL methodology.
Author{1}{Firstname}#=%=#Huixin
Author{1}{Lastname}#=%=#Chen
Author{1}{Username}#=%=#chenhuixin
Author{1}{Email}#=%=#huixin333666@gmail.com
Author{1}{Affiliation}#=%=#Ludwig Maximilian University of Munich
Author{2}{Firstname}#=%=#Jan
Author{2}{Lastname}#=%=#Büssing
Author{2}{Email}#=%=#jan.buessing@campus.lmu.de
Author{2}{Affiliation}#=%=#Institute for Statistics, LMU Munich
Author{3}{Firstname}#=%=#David
Author{3}{Lastname}#=%=#Rügamer
Author{3}{Email}#=%=#david.ruegamer@stat.uni-muenchen.de
Author{3}{Affiliation}#=%=#Institute for Statistics, LMU Munich
Author{4}{Firstname}#=%=#Ercong
Author{4}{Lastname}#=%=#Nie
Author{4}{Username}#=%=#ercong
Author{4}{Email}#=%=#nie@cis.lmu.de
Author{4}{Affiliation}#=%=#Centre for Information and Language Processing, LMU Munich

==========
èéáğö