MentalHospital, a virtual evaluation environment for LLM-based psychiatric clinical encounters, and MentalEval, five domain-specific evaluators covering communication empathy, interviewing professionalism, clinical-note quality, diagnostic rigor, and treatment appropriateness, trained with rubric-grounded SFT and expert-guided DPO are introduced.
Yu-Ming Yang, Xiao Sun, Yuanwei Zou et al.· arXiv.org· 0 citations
GeoForge is a training-free, self-evolving framework that transforms completed trajectories into a structured nonparametric execution state, and consistently improves both task accuracy and tool-use trajectory quality across diverse LLM backbones, while substantially reducing tool-planning and reasoning errors for most LLMs.
Xin Xiao, Jiang Zhong, Junnan Zhu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.