Deep Learning Mental Health Dialogue System

TL;DR

This paper introduces Serena, a deep learning-based mental health dialogue system that aims to improve access to counseling by providing empathetic, engaging responses, while addressing challenges like coherence and hallucination.

Contribution

The paper presents a novel large-scale transformer-based dialogue system for mental health support, including post-processing algorithms to enhance response quality.

Findings

Serena can generate empathetic and engaging responses.

The system demonstrates potential as a low-cost supplement to human counseling.

Challenges like hallucination and incoherence are identified and partially addressed.

Abstract

Mental health counseling remains a major challenge in modern society due to cost, stigma, fear, and unavailability. We posit that generative artificial intelligence (AI) models designed for mental health counseling could help improve outcomes by lowering barriers to access. To this end, we have developed a deep learning (DL) dialogue system called Serena. The system consists of a core generative model and post-processing algorithms. The core generative model is a 2.7 billion parameter Seq2Seq Transformer fine-tuned on thousands of transcripts of person-centered-therapy (PCT) sessions. The series of post-processing algorithms detects contradictions, improves coherency, and removes repetitive answers. Serena is implemented and deployed on \url{https://serena.chat}, which currently offers limited free services. While the dialogue system is capable of responding in a qualitatively empathetic and engaging manner, occasionally it displays hallucination and long-term incoherence. Overall, we demonstrate that a deep learning mental health dialogue system has the potential to provide a low-cost and effective complement to traditional human counselors with less barriers to access.