Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Mixed Chain-of-Psychotherapies for Emotional Support Chatbot

Siyuan Chen, Cong Ming, Zhiling Zhang, Yanyi Chen, Kenny Q. Zhu, Mengyue Wu

TL;DR

PsyMix is a chatbot that integrates the analyses of the seeker's state from the perspective of a psychotherapy approach before generating the response, and learns to incorporate the strength of various psychotherapies by fine-tuning on a mixture of CoPs.

Abstract

In the realm of mental health support chatbots, it is vital to show empathy and encourage self-exploration to provide tailored solutions. However, current approaches tend to provide general insights or solutions without fully understanding the help-seeker's situation. Therefore, we propose PsyMix, a chatbot that integrates the analyses of the seeker's state from the perspective of a psychotherapy approach (Chain-of-Psychotherapies, CoP) before generating the response, and learns to incorporate the strength of various psychotherapies by fine-tuning on a mixture of CoPs. Through comprehensive evaluation, we found that PsyMix can outperform the ChatGPT baseline, and demonstrate a comparable level of empathy in its responses to that of human counselors.

Mixed Chain-of-Psychotherapies for Emotional Support Chatbot

TL;DR

PsyMix is a chatbot that integrates the analyses of the seeker's state from the perspective of a psychotherapy approach before generating the response, and learns to incorporate the strength of various psychotherapies by fine-tuning on a mixture of CoPs.

Abstract

In the realm of mental health support chatbots, it is vital to show empathy and encourage self-exploration to provide tailored solutions. However, current approaches tend to provide general insights or solutions without fully understanding the help-seeker's situation. Therefore, we propose PsyMix, a chatbot that integrates the analyses of the seeker's state from the perspective of a psychotherapy approach (Chain-of-Psychotherapies, CoP) before generating the response, and learns to incorporate the strength of various psychotherapies by fine-tuning on a mixture of CoPs. Through comprehensive evaluation, we found that PsyMix can outperform the ChatGPT baseline, and demonstrate a comparable level of empathy in its responses to that of human counselors.
Paper Structure (27 sections, 5 figures, 8 tables)

This paper contains 27 sections, 5 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Psychotherapy Approaches
  3. Method
  4. Chain-of-Psychotherapies (CoP)
  5. SFT with Mixed CoP
  6. Experiments
  7. Dataset
  8. Experiment Settings
  9. Methods for Comparison
  10. Human Evaluation
  11. Evaluation Procedure
  12. Results
  13. Empathy Analysis
  14. Results
  15. Case Study
  16. ...and 12 more sections

Figures (5)

  • Figure 1: Comparison of the responses generated by ChatGPT and PsyMix based on the same context.
  • Figure 2: Illustration of the two-step pipeline. We first generate CoPs via ChatGPT, then pack CoP with the original utterance (i.e., ground truth) in the dialog for supervised fine-tuning.
  • Figure 3: Chain-of-Psychotherapies for case 1.
  • Figure 4: Chain-of-Psychotherapies for case 2.
  • Figure 5: Chain-of-Psychotherapies for case 3.