Large Language Models for Mental Health Interventions

We study the empirical and regulatory foundations needed to deploy large language models (LLMs) safely in mental health care. We conducted among the first real-world evaluations of a conversational AI agent built for anxiety and depression, and we design simulated patients to stress-test conversational systems before they reach people. To guide development, we authored a research framework for LLM-based interventions and an implementation framework covering regulatory considerations, ethical safeguards, and deployment barriers specific to mental health.

Selected Publications

Wolfe, E. C., Su, T., Tieleman, O., Hull, T. D., Malgaroli, M., Stamatis, C. A. (2026). Engagement phenotypes for a sample of 102,684 AI mental health chatbot users and dose-response associations with clinical outcomes. arXiv preprint.

Language ModelsHuman–AIComputationalDigital Health

Beikzadeh, M., Asadollah Salmanpour, Y., Suvarna, A., Sankararaman, S., Malgaroli, M., Sarrafzadeh, M., Gabriel, S. (2026). Multi-objective alignment of language models for personalized psychotherapy. arXiv preprint.

Language Models

Jin, K. W., Rostam-Abadi, Y., Chaudhary, P., Garrett, M. A., Huang, A. S., Montelongo, M., Nagpal, C., Shei, J., Weathers, J., Zhang, J. S., Chen, Q., Kim, J., Malgaroli, M., Mathis, W. S., Rodriguez, C. I., Selek, S., Sharma, M. S., Pittenger, C., Yip, S. W., Zaboski, B. A., Xu, H. (2026). Evaluating diagnostic accuracy and clinical reasoning of multiple large language models in psychiatry. medRxiv preprint.

NLP & Language MarkersLanguage Models

Stamatis, C. A., Meyerhoff, J., Zhang, R., Tieleman, O., Malgaroli, M., Hull, T. D. (2026). Beyond simulations: What 20,000 real conversations reveal about mental health AI safety. arXiv preprint.

Language ModelsHuman–AI

Hull, T. D., Zhang, L., Arean, P. A., Malgaroli, M. (2025). Generative AI purpose-built for social and mental health: A real-world pilot. arXiv preprint.

Language ModelsHuman–AIDigital Health

Zhu, Z., Tieleman, O., Stamatis, C. A., Smyth, L., Hull, T. D., Cahn, D. R., Chen, J., Malgaroli, M. (2026). DIAL: Direct Iterative Adversarial Learning for Realistic Multi-Turn Dialogue Simulation. Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP).

NLP & Language MarkersLanguage ModelsHuman–AI

Gabriel, S., Puri, I., Xu, X., Malgaroli, M., Ghassemi, M. (2024). Can AI relate: Testing large language model response for mental health support. Findings of the Association for Computational Linguistics: EMNLP 2024.

NLP & Language MarkersLanguage ModelsHuman–AIComputational

Galatzer-Levy, I. R., McDuff, D., Natarajan, V., Karthikesalingam, A., Malgaroli, M. (2023). The capability of large language models to measure psychiatric functioning. arXiv.

NLP & Language MarkersLanguage Models

Galatzer-Levy, I., McDuff, D., Malgaroli, M. (2025). The Capability of Large Language Models to Measure and Differentiate Psychiatric Conditions Through O-Shot Learning. Biological Psychiatry.

NLP & Language MarkersLanguage Models

Malgaroli, M., Robinaugh, D., Bonanno, G., Simon, N. (2025). Leveraging Large Language Models and Phonetic Analysis to Predict Adjustment to Bereavement and Social Disruptions. Biological Psychiatry.

NLP & Language MarkersLanguage Models

Malgaroli, M., Schultebraucks, K., Myrick, K. J., Loch, A. A., Ospina-Pinillos, L., Choudhury, T., Kotov, R., De Choudhury, M., Torous, J. (2025). Large language models for the mental health community: Framework for translating code to care. The Lancet Digital Health.

Language ModelsDigital Health

Malgaroli, M., McDuff, D. (2024). An overview of diagnostics and therapeutics using large language models. Journal of Traumatic Stress.

NLP & Language MarkersLanguage Models

See all publications →