Does Rationale Quality Matter? Enhancing Mental Disorder Detection via Selective Reasoning Distillation

TL;DR

This paper explores how high-quality, domain-relevant rationales improve the performance of smaller language models in detecting mental health issues from social media, emphasizing the importance of rationale quality in model distillation.

Contribution

It introduces a framework for selecting rationales based on clinical relevance to enhance mental health detection and explanation in smaller language models.

Findings

Quality-focused rationale selection improves detection accuracy.

Enhanced rationale relevance leads to better explanation generation.

The approach outperforms baseline models in mental health tasks.

Abstract

The detection of mental health problems from social media and the interpretation of these results have been extensively explored. Research has shown that incorporating clinical symptom information into a model enhances domain expertise, improving its detection and interpretation performance. While large language models (LLMs) are shown to be effective for generating explanatory rationales in mental health detection, their substantially large parameter size and high computational cost limit their practicality. Reasoning distillation transfers this ability to smaller language models (SLMs), but inconsistencies in the relevance and domain alignment of LLM-generated rationales pose a challenge. This paper investigates how rationale quality impacts SLM performance in mental health detection and explanation generation. We hypothesize that ensuring high-quality and domain-relevant rationales enhances the distillation. To this end, we propose a framework that selects rationales based on their alignment with expert clinical reasoning. Experiments show that our quality-focused approach significantly enhances SLM performance in both mental disorder detection and rationale generation. This work highlights the importance of rationale quality and offers an insightful framework for knowledge transfer in mental health applications.