SubmissionNumber#=%=#120
FinalPaperTitle#=%=#AAdaM at SemEval-2024 Task 1: Augmentation and Adaptation for Multilingual Semantic Textual Relatedness
ShortPaperTitle#=%=#
NumberOfPages#=%=#11
CopyrightSigned#=%=#Miaoran Zhang
JobTitle#==#
Organization#==#
Abstract#==#This paper presents our system developed for the SemEval-2024 Task 1: Semantic Textual Relatedness for African and Asian Languages. The shared task aims at measuring the semantic textual relatedness between pairs of sentences, with a focus on a range of under-represented languages. In this work, we propose using machine translation for data augmentation to address the low-resource challenge of limited training data. Moreover, we apply task-adaptive pre-training on unlabeled task data to bridge the gap between pre-training and task adaptation. For model training, we investigate both full fine-tuning and adapter-based tuning, and adopt the adapter framework for effective zero-shot cross-lingual transfer. We achieve competitive results in the shared task: our system performs the best among all ranked teams in both subtask A (supervised learning) and subtask C (cross-lingual transfer).
Author{1}{Firstname}#=%=#Miaoran
Author{1}{Lastname}#=%=#Zhang
Author{1}{Username}#=%=#mzhang_1
Author{1}{Email}#=%=#mzhang@lsv.uni-saarland.de
Author{1}{Affiliation}#=%=#Saarland University
Author{2}{Firstname}#=%=#Mingyang
Author{2}{Lastname}#=%=#Wang
Author{2}{Username}#=%=#my26w
Author{2}{Email}#=%=#mingyang.wang2@de.bosch.com
Author{2}{Affiliation}#=%=#Bosch Center for Artificial Intelligence; LMU Munich
Author{3}{Firstname}#=%=#Jesujoba
Author{3}{Lastname}#=%=#Alabi
Author{3}{Username}#=%=#alabijesujoba
Author{3}{Email}#=%=#s8jealab@stud.uni-saarland.de
Author{3}{Affiliation}#=%=#Saarland University
Author{4}{Firstname}#=%=#Dietrich
Author{4}{Lastname}#=%=#Klakow
Author{4}{Username}#=%=#dietrich.klakow
Author{4}{Email}#=%=#dklakow@lsv.uni-saarland.de
Author{4}{Affiliation}#=%=#Saarland University

==========
èéáğö