HU at SemEval-2024 Task 8A: Can Contrastive Learning Learn Embeddings to Detect Machine-Generated Text?

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dipta, Shubhashis Roy, Shahriar, Sadat
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910387852541952
author Dipta, Shubhashis Roy
Shahriar, Sadat
author_facet Dipta, Shubhashis Roy
Shahriar, Sadat
contents This paper describes our system developed for SemEval-2024 Task 8, ``Multigenerator, Multidomain, and Multilingual Black-Box Machine-Generated Text Detection'' Machine-generated texts have been one of the main concerns due to the use of large language models (LLM) in fake text generation, phishing, cheating in exams, or even plagiarizing copyright materials. A lot of systems have been developed to detect machine-generated text. Nonetheless, the majority of these systems rely on the text-generating model. This limitation is impractical in real-world scenarios, as it's often impossible to know which specific model the user has used for text generation. In this work, we propose a $\textbf{single}$ model based on contrastive learning, which uses $\textbf{$\approx$40% of the baseline's parameters}$ (149M vs. 355M) but shows a comparable performance on the test dataset $(\textbf{21st out of 137 participants})$. Our key finding is that even without an ensemble of multiple models, a single base model can have comparable performance with the help of data augmentation and contrastive learning. Our code is publicly available at https://github.com/dipta007/SemEval24-Task8.
format Preprint
id arxiv_https___arxiv_org_abs_2402_11815
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle HU at SemEval-2024 Task 8A: Can Contrastive Learning Learn Embeddings to Detect Machine-Generated Text?
Dipta, Shubhashis Roy
Shahriar, Sadat
Computation and Language
Artificial Intelligence
Machine Learning
This paper describes our system developed for SemEval-2024 Task 8, ``Multigenerator, Multidomain, and Multilingual Black-Box Machine-Generated Text Detection'' Machine-generated texts have been one of the main concerns due to the use of large language models (LLM) in fake text generation, phishing, cheating in exams, or even plagiarizing copyright materials. A lot of systems have been developed to detect machine-generated text. Nonetheless, the majority of these systems rely on the text-generating model. This limitation is impractical in real-world scenarios, as it's often impossible to know which specific model the user has used for text generation. In this work, we propose a $\textbf{single}$ model based on contrastive learning, which uses $\textbf{$\approx$40% of the baseline's parameters}$ (149M vs. 355M) but shows a comparable performance on the test dataset $(\textbf{21st out of 137 participants})$. Our key finding is that even without an ensemble of multiple models, a single base model can have comparable performance with the help of data augmentation and contrastive learning. Our code is publicly available at https://github.com/dipta007/SemEval24-Task8.
title HU at SemEval-2024 Task 8A: Can Contrastive Learning Learn Embeddings to Detect Machine-Generated Text?
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2402.11815