This paper compares the effectiveness of traditional machine learning methods, encoder-based models, and large language models (LLMs) on the task of detecting depression and anxiety. Five Russian-language datasets were considered, each differing in format and in the method used to define the target pathology class. We tested AutoML models based on linguistic features, several variations of encoder-based Transformers such as BERT, and state-of-the-art LLMs as pathology classification models. The results demonstrated that LLMs outperform traditional methods, particularly on noisy and small datasets where training examples vary significantly in text length and genre. However, psycholinguistic features and encoder-based models can achieve performance comparable to language models when trained on texts from individuals with clinically confirmed depression, highlighting their potential effectiveness in targeted clinical applications.
DOI: 10.18653/v1/2025.emnlp-main.1752
Download the article (PDF) at ACL Anthology: https://aclanthology.org/2025.emnlp-main.1752/
Download the article (PDF) or read online at arXiv.org: https://arxiv.org/html/2410.07129
eLibrary: https://elibrary.ru/item.asp?id=91847222
ResearchGate: https://www.researchgate.net/publication/397421223_Exploring_Large_Language_Models_for_Detecting_Mental_Disorders
Kuzmin G., Strepetov P., Stankevich M, Chudova N, Shelmanov A., Smirnov I. Exploring Large Language Models for Detecting Mental Disorders // Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP 2025), 2025, pp. 34535–34559.