CPEMH: An Agentic Framework for Prompt-Driven Behavior Evaluation and Assurance in Foundation-Model Systems for Mental Health Screening
For developers of clinical AI systems, CPEMH provides a methodology to control behavioral variability in sensitive domains, though it is an incremental engineering approach rather than a fundamental advance.
CPEMH is an agentic framework for evaluating and assuring prompt-driven behavior in foundation-model systems for mental health screening. In a depression screening case study, it stabilizes model behavior and enables systematic auditing, prioritizing stability over architectural complexity.
This paper presents CPEMH, an agentic framework designed to evaluate prompt-driven behavior in foundation-model systems operating on transcript-based datasets for mental-health screening. CPEMH serves as an engineering methodology for behavioral assurance in large-scale language systems, introducing an orchestrated architecture that autonomously performs the design, evaluation, and selection of prompt strategies, enabling systematic control of behavioral variability across contexts. Its modular agentic design, combining orchestrator, inference, and evaluation agents, ensures traceability, reproducibility, and robustness throughout the prompting lifecycle. A case study on automated depression screening from interview transcripts demonstrates the framework's capacity to stabilize and audit foundation-model behavior in conversational and clinically sensitive domains. Lessons learned emphasize the role of modular orchestration in behavioral assurance, the prioritization of stability over architectural complexity, and the integration of F1, bias, and robustness as core acceptance criteria.