aceOptimizing Small Local Language Models for Culturally Competent Mental Health Counseling: Comparative Evaluation With GPT-4o by Psychiatrists
Background: Large language models (LLMs) show promise for mental health counseling applications. However, their deployment faces significant challenges, including data privacy concerns requiring on-premise solutions, high operational costs, and Western-centric biases that limit cultural appropriateness for non-English-speaking populations. Objective: This study aimed to develop and evaluate culturally competent small language models (SLMs) with approximately 7 to 8 billion parameters for Korean mental health cou…
A fine-tuned 7B Korean small language model showed no statistically significant difference from GPT-4o in supportiveness and cultural appropriateness, as rated by psychiatrists, supporting local, privacy-preserving deployment.
All tested small language models scored significantly lower than GPT-4o on safety in mental health counseling scenarios, so human expert oversight is still needed.
The safety gap versus GPT-4o means the small models need human expert oversight, and the LLM-as-a-Judge showed a ceiling effect in safety scoring, limiting automated screening for safety.
Evidence
- Peer-reviewedJMIR AI2026-10-22
How should this claim be treated?
Truvace Impact Record TRV-2026-2247, v1: “Optimizing Small Local Language Models for Culturally Competent Mental Health Counseling: Comparative Evaluation With GPT-4o by Psychiatrists.” Truvace, 2026-10-25. /record/TRV-2026-2247 (accessed at citation time). sha256 91b18decb510a3e8…
Calibration history
Every change to this record since certification, in the open. None yet — the reading has held since it entered the record.
Certified into the record
How to verify without trusting this page
Fetch the canonical text of any version from /api/record/TRV-2026-2247 and hash it yourself — for example shasum -a 256 on the saved canonical field. The result must equal content_hash, and each version’s text ends with prev:followed by the prior version’s hash (version 1 chains to 64 zeros). If a single character of any version had been altered since certification, the chain would not reproduce.