Enhancing Sentiment Classification and Irony Detection in Large Language Models through Advanced Prompt Engineering Techniques
It addresses improving sentiment analysis for users of LLMs, but is incremental as it applies known prompting methods to specific models and tasks.
This study tackled sentiment analysis tasks like classification and irony detection using advanced prompt engineering on LLMs such as GPT-4o-mini and gemini-1.5-flash, finding that techniques like few-shot and chain-of-thought prompting significantly improved performance, with up to a 46% boost in irony detection for gemini-1.5-flash.
This study investigates the use of prompt engineering to enhance large language models (LLMs), specifically GPT-4o-mini and gemini-1.5-flash, in sentiment analysis tasks. It evaluates advanced prompting techniques like few-shot learning, chain-of-thought prompting, and self-consistency against a baseline. Key tasks include sentiment classification, aspect-based sentiment analysis, and detecting subtle nuances such as irony. The research details the theoretical background, datasets, and methods used, assessing performance of LLMs as measured by accuracy, recall, precision, and F1 score. Findings reveal that advanced prompting significantly improves sentiment analysis, with the few-shot approach excelling in GPT-4o-mini and chain-of-thought prompting boosting irony detection in gemini-1.5-flash by up to 46%. Thus, while advanced prompting techniques overall improve performance, the fact that few-shot prompting works best for GPT-4o-mini and chain-of-thought excels in gemini-1.5-flash for irony detection suggests that prompting strategies must be tailored to both the model and the task. This highlights the importance of aligning prompt design with both the LLM's architecture and the semantic complexity of the task.