# Conversation Model Fine-Tuning for Classifying Client Utterances in Counseling Dialogues

arXiv: https://arxiv.org/abs/1904.00350 · HTML: https://mlsanity.com/paper/1904.00350/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Mar 2019.

## Built on by (most cited first, 8 of 8)

- [Understanding Client Reactions in Online Mental Health Counseling](https://mlsanity.com/paper/2306.15334/) (2023)
- [Taking a turn for the better: Conversation redirection throughout the course of mental-health therapy](https://mlsanity.com/paper/2410.07147/) (2024)
- [Towards Automated Real-time Evaluation in Text-based Counseling](https://mlsanity.com/paper/2203.03442/) (2022)
- [RECAP: Resistance Capture in Text-based Mental Health Counseling with Large Language Models](https://mlsanity.com/paper/2601.14780/) (2026)
- [Towards Understanding Counseling Conversations: Domain Knowledge and Large Language Models](https://mlsanity.com/paper/2402.14200/) (2024)
- [Evaluating Role-Consistency in LLMs for Counselor Training](https://mlsanity.com/paper/2601.08892/) (2026)
- [Automated Utterance Labeling of Conversations Using Natural Language Processing](https://mlsanity.com/paper/2208.06525/) (2022)
- [Transition to Adulthood for Young People with Intellectual or Developmental Disabilities: Emotion Detection and Topic Modeling](https://mlsanity.com/paper/2209.10477/) (2022)

## Abstract (arXiv)

The recent surge of text-based online counseling applications enables us to collect and analyze interactions between counselors and clients. A dataset of those interactions can be used to learn to automatically classify the client utterances into categories that help counselors in diagnosing client status and predicting counseling outcome. With proper anonymization, we collect counselor-client dialogues, define meaningful categories of client utterances with professional counselors, and develop a novel neural network model for classifying the client utterances. The central idea of our model, ConvMFiT, is a pre-trained conversation model which consists of a general language model built from an out-of-domain corpus and two role-specific language models built from unlabeled in-domain dialogues. The classification result shows that ConvMFiT outperforms state-of-the-art comparison models. Further, the attention weights in the learned model confirm that the model finds expected linguistic patterns for each category.