P-ReMIS: Pragmatic Reasoning in Mental Health and a Social Implication

TL;DR

This paper explores the reasoning capabilities of large language models in mental health, introducing a new dataset and tasks to evaluate pragmatic reasoning, and examines models' behavior on stigma-related issues.

Contribution

It introduces the PRiMH dataset and pragmatic reasoning tasks in mental health, benchmarking multiple models and proposing prompts to study stigma in LLMs.

Findings

Mistral and Qwen show strong reasoning abilities in mental health tasks.

Claude-3.5-haiku handles mental health stigma more responsibly.

Proposed prompts effectively evaluate stigma awareness in LLMs.

Abstract

Although explainability and interpretability have received significant attention in artificial intelligence (AI) and natural language processing (NLP) for mental health, reasoning has not been examined in the same depth. Addressing this gap is essential to bridge NLP and mental health through interpretable and reasoning-capable AI systems. To this end, we investigate the pragmatic reasoning capability of large-language models (LLMs) in the mental health domain. We introduce PRiMH dataset, and propose pragmatic reasoning tasks in mental health with pragmatic implicature and presupposition phenomena. In particular, we formulate two tasks in implicature and one task in presupposition. To benchmark the dataset and the tasks presented, we consider four models: Llama3.1, Mistral, MentaLLaMa, and Qwen. The results of the experiments suggest that Mistral and Qwen show substantial reasoning abilities in the domain. Subsequently, we study the behavior of MentaLLaMA on the proposed reasoning tasks with the rollout attention mechanism. In addition, we also propose three StiPRompts to study the stigma around mental health with the state-of-the-art LLMs, GPT4o-mini, Deepseek-chat, and Claude-3.5-haiku. Our evaluated findings show that Claude-3.5-haiku deals with stigma more responsibly compared to the other two LLMs.