Prompt sensitivity of transformer variants for text classification

• المؤلف الرئيسي: Ong, Li Han
• مؤلفون آخرون: Wang Wenya
• التنسيق: Final Year Project
• اللغة: English
• منشور في: Nanyang Technological University 2024
• الموضوعات: Computer and Information Science; Machine learning; Classification; Model sensitivity; Large language models
• الوصول للمادة أونلاين: https://hdl.handle.net/10356/181519
• المؤسسة: Nanyang Technological University

This study investigates the sensitivity of various Transformer model architectures, encoder-only (BERT), decoder-only (GPT-2), and encoder-decoder (T5), in response to various types of prompt modifications on text classification tasks. By leveraging a fine-tuning approach, the models were evaluated...