An Initial Exploration of Contrastive Prompt Tuning to Generate Energy-Efficient Code

Researchers explored using Contrastive Prompt Tuning (CPT) to improve Large Language Models' ability to generate energy-efficient code, combining contrastive learning with parameter-efficient fine-tuning. The study tested CPT across Python, Java, and C++ on three different models, finding consistent accuracy improvements for two models but variable efficiency gains depending on model, language, and task complexity.

• →LLMs typically generate functionally correct but energy-inefficient code compared to human-written solutions.
• →Contrastive Prompt Tuning combines contrastive learning techniques with parameter-efficient fine-tuning to distinguish between efficient and inefficient code.
• →The method was tested across Python, Java, and C++ programming languages on three different AI models.
• →Results showed consistent code accuracy improvements for two models but efficiency gains varied significantly.
• →The approach supports Green Software Development efforts aimed at reducing computational energy consumption.