The Flow has not summarised this story yet — read it at arXiv Computation and Language.
arXiv:2608.29010v1 Announce Type: cross Abstract: People share mental health diagnoses on social media, yet how such language becomes visible around their self-disclosure, and whether community engag...
arXiv:2606. 19640v1 Announce Type: cross Abstract: AI and large language models (LLMs) have emerged as promising tools to address global mental health challenges.
The study analyzes how mental health self‑disclosure on Twitter changed after Elon Musk’s 2022 acquisition, examining 824K tweets related to anxiety, depression, bipolar disorder, borderline personality disorder, and suicidal ideation. Using LIWC and an LLM‑based intimacy classifier, researchers found a post‑acquisition decline in affective and social language, a rise in first‑person singular pronouns and future‑focused expressions, and varied intimacy patterns across conditions—low‑intimacy disclosures increased for anxiety, depression, and BPD, while high‑intimacy disclosures rose for suicide‑related posts. Only the moderate‑versus‑low intimacy decline for BPD remained statistically significant compared to broader Twitter discourse.
The paper reviews how large language models are applied in mental health, covering areas such as social media analysis, clinical conversational agents, therapy support tools, prompt engineering, and multimodal learning. It synthesizes interdisciplinary studies that use social media posts, electronic medical records, and multimodal inputs to detect depression, assess suicide risk, provide personalized therapy, and generate psychoeducational content. The review also discusses advances in model interpretability, annotation strategies, multimodal fusion techniques, and highlights ethical, sociotechnical, and regulatory challenges while proposing frameworks for safe, equitable, and accountable deployment.
arXiv:2606. 27247v1 Announce Type: new Abstract: In NLP, mental health conditions are often modeled as isolated phenomena, without interpersonal context.
arXiv:2607. 05685v1 Announce Type: cross Abstract: Large language models are increasingly used as private, always-available conversational systems, but little is known about how people with depressive symptoms use them.