Deng, X. Privacy Preserving Language Models Detect Early Warning Signs in Digital Mental Health. JCEIM 2026, 21 (2), 18-24. https://doi.org/10.54097/d4bfqr62.