Spam Email Classification using LoRA Fine-tuned Transformers: High-performance spam email classification using LoRA-adapted transformer models (ELECTRA, RoBERTa). Achieves 99.4%+ accuracy with parameter-efficient fine-tuning on 83K+ emails.
nlp deep-learning text-classification transformers pytorch lora spam-detection spam-classification peft huggingface ablation-study huggingface-transformers llm email-filtering llm-finetuning model-finetuning imbalanced-dataset-handling parameter-efficient-finetuning lora-finetuning bert-based-models
-
Updated
Mar 12, 2026 - Jupyter Notebook