Enhancing AI-Driven Psychological Consultation: Layered Prompts with Large Language Models

TL;DR

This paper presents a layered prompting system for large language models to improve AI-driven psychological consultations by enhancing emotional understanding and adaptability, validated through experiments showing improved response quality.

Contribution

It introduces a novel layered prompt architecture combined with empathy and scenario prompts to enhance LLMs' therapeutic capabilities in mental health support.

Findings

Significant improvement in response relevance and empathy

Enhanced adaptability to user input

Potential for scalable mental health support

Abstract

Psychological consultation is essential for improving mental health and well-being, yet challenges such as the shortage of qualified professionals and scalability issues limit its accessibility. To address these challenges, we explore the use of large language models (LLMs) like GPT-4 to augment psychological consultation services. Our approach introduces a novel layered prompting system that dynamically adapts to user input, enabling comprehensive and relevant information gathering. We also develop empathy-driven and scenario-based prompts to enhance the LLM's emotional intelligence and contextual understanding in therapeutic settings. We validated our approach through experiments using a newly collected dataset of psychological consultation dialogues, demonstrating significant improvements in response quality. The results highlight the potential of our prompt engineering techniques to enhance AI-driven psychological consultation, offering a scalable and accessible solution to meet the growing demand for mental health support.