Are LLMs Effective Backbones for Fine-tuning? An Experimental Investigation of Supervised LLMs on Chinese Short Text Matching
This work addresses the problem of adapting LLMs for specific supervised tasks in natural language understanding, particularly for Chinese text, but it appears incremental as it builds on existing fine-tuning methods without introducing a new paradigm.
The study investigates the effectiveness of fine-tuning large language models (LLMs) for Chinese short text matching, exploring factors like task modeling and prompt formats to assess performance in supervised settings.
The recent success of Large Language Models (LLMs) has garnered significant attention in both academia and industry. Prior research on LLMs has primarily focused on enhancing or leveraging their generalization capabilities in zero- and few-shot settings. However, there has been limited investigation into effectively fine-tuning LLMs for a specific natural language understanding task in supervised settings. In this study, we conduct an experimental analysis by fine-tuning LLMs for the task of Chinese short text matching. We explore various factors that influence performance when fine-tuning LLMs, including task modeling methods, prompt formats, and output formats.