A systematic review of large language models (LLMs) in mental health applications has been published. The study covers various uses of LLMs, including social media analysis, clinical conversational agents, and therapy support tools. It highlights advancements in model interpretability and annotation strategies, as well as the importance of prompt engineering for domain adaptation. The review also discusses emerging multimodal fusion techniques that integrate text, speech, and sensor data for improved mental health diagnosis and monitoring. However, it emphasizes ongoing ethical challenges, such as ensuring safe and equitable deployment of LLMs in real-world care. --- Why it matters: This study matters to AI researchers because it highlights the potential applications of large language models in mental health diagnosis and treatment, while also emphasizing the need for careful consideration of ethical implications. Source: https://arxiv.org/abs/2608.18080