Reproduce “Next-Generation Spam Filtering: Comparative Fine-Tuning of LLMs, NLPs, and CNN Models for Email Spam Classification” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Konstantinos I. Roumeliotis, Nikolaos D. Tselikas, Dimitrios K. Nasiopoulos
The authors compare fine-tuned GPT-4, BERT, and RoBERTa with a trained convolutional neural network on two labeled spam-and-ham text datasets. The models classify message text as spam or ham, and the study tests performance both within each dataset and across them. BERT reaches 99.39% accuracy on one held-out test set; these results apply to simplified benchmark text, not a proven phishing defense.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.