The Empirically Grounded Adaptive Virtual Patient for Psychotherapy Training: Disclosure That Responds to Therapist Micro-Skills

This RSS feed aggregates new arXiv CS/CY submissions (10 Jun 2026) covering several AI-society and ML-systems topics: (1) Adaptive Virtual Patient — an LLM-driven psychotherapy simulator whose disclosure dynamics are grounded in a structural equation model fit to ~2,000 hours of therapy transcripts to adapt patient openness to trainee micro-skills; (2) Trustworthiness ideals for AI-powered peripartum information — focus groups highlight the need for inspectable transparency, recourse, and ecosystem-aware governance in high-stakes maternal health; (3) AnnotateThis — a human–LLM annotation/LLM‑g