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.

What problem does this paper attempt to address?

The main problem this paper attempts to address is the challenge of accessibility and scalability in psychological counseling services. Specifically: 1. **Shortage of qualified professionals**: Psychological counseling services require a large number of qualified mental health professionals, but in many areas, especially remote or underdeveloped regions, such resources are very scarce. 2. **Scalability of services**: Traditional face-to-face counseling models struggle to meet the growing demand for mental health services, particularly when it comes to large-scale implementation. 3. **Privacy and confidentiality**: Online psychological counseling services need to ensure user privacy and information confidentiality, which adds to the complexity of the service. To address these issues, the paper proposes a new approach based on large language models (such as GPT-4) by introducing a layered prompting system to enhance the effectiveness of psychological counseling services. This approach includes the following aspects: - **Layered prompting system that dynamically adapts to user input**: Through a series of dynamic prompts, the system can gradually collect detailed information from users, thereby providing more comprehensive and relevant information. - **Emotion-driven prompts**: These prompts encourage the model to respond to users in a compassionate and supportive manner, enhancing the model's emotional intelligence. - **Context-based prompts**: By providing specific situational cases, the model is trained to better understand and simulate real-life counseling scenarios. The paper validates the effectiveness of this approach through experiments, using a newly collected psychological counseling dialogue dataset for evaluation. The results show significant improvements in response quality, empathy, and contextual understanding. These improvements help enhance the quality and accessibility of AI-driven psychological counseling services, meeting the growing demand for mental health support.

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

Toward Large Language Models as a Therapeutic Tool: Comparing Prompting Techniques to Improve GPT-Delivered Problem-Solving Therapy

Can AI Relate: Testing Large Language Model Response for Mental Health Support

Harnessing Large Language Models' Empathetic Response Generation Capabilities for Online Mental Health Counselling Support

Psy-LLM: Scaling up Global Mental Health Psychological Services with AI-based Large Language Models

Supporting the Demand on Mental Health Services with AI-Based Conversational Large Language Models (LLMs)

Optimizing Psychological Counseling with Instruction-Tuned Large Language Models

Can Large Language Models be Used to Provide Psychological Counselling? An Analysis of GPT-4-Generated Responses Using Role-play Dialogues

Response Generation for Cognitive Behavioral Therapy with Large Language Models: Comparative Study with Socratic Questioning

Advancing Conversational Psychotherapy: Integrating Privacy, Dual-Memory, and Domain Expertise with Large Language Models

Empowering Psychotherapy with Large Language Models: Cognitive Distortion Detection through Diagnosis of Thought Prompting

EmotionPrompt: Leveraging Psychology for Large Language Models Enhancement via Emotional Stimulus

Enhancing Psychological Counseling with Large Language Model: A Multifaceted Decision-Support System for Non-Professionals

WundtGPT: Shaping Large Language Models To Be An Empathetic, Proactive Psychologist

Harnessing the Power of Large Language Models for Empathetic Response Generation: Empirical Investigations and Improvements

Data Augmentation of Multi-turn Psychological Dialogue via Knowledge-driven Progressive Thought Prompting

Enhancing Mental Health Support through Human-AI Collaboration: Toward Secure and Empathetic AI-enabled chatbots

Task Supportive and Personalized Human-Large Language Model Interaction: A User Study

ChatGPT-HealthPrompt. Harnessing the Power of XAI in Prompt-Based Healthcare Decision Support using ChatGPT

Domain-Specific Improvement on Psychotherapy Chatbot Using Assistant