DistilBERT: A Novel Approach to Detect Text Generated by Large Language Models (LLM)
Preprint
- 1 February 2024
- preprint
- Published by Springer Nature in Research Square
Abstract
Large language models (LLMs) have emerged as powerful tools for generating human-quality text, raising concerns about their potential for misuse in academic settings. This paper investigates the use of DistilBERT, a distilled version of BERT, for detecting LLM-generated text. We evaluate its performance on two publicly available datasets, LLM-Detect AI Generated Text and DAIGT-V3 Train Dataset, achieving an average accuracy of around 94%. Our findings suggest that DistilBERT is a promising tool for safeguarding academic integrity in the era of LLMs.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- ChatGPT Isn't MagicM/C Journal, 2023
- Exploring the Ethical Dimensions of Using ChatGPT in Language Learning and BeyondLanguages, 2023
- The effectiveness of T5, GPT-2, and BERT on text-to-image generation taskPattern Recognition Letters, 2023
- ChatGPT and large language model (LLM) chatbots: The current state of acceptability and a proposal for guidelines on utilization in academic medicineJournal of Pediatric Urology, 2023
- Using ChatGPT and Other Large Language Model (LLM) Applications for Academic Paper AssignmentsSocArXiv, 2023
- ChatGPT is fun, but not an authorScience, 2023
- Analyzing the Performance of Sentiment Analysis using BERT, DistilBERT, and RoBERTa2022
- Natural language processing (NLP) in management research: A literature reviewJournal of Management Analytics, 2020
- How to Fine-Tune BERT for Text Classification?2019
- A texture classifier based on neural network principles1990