CPEMH: An Agentic Framework for Prompt-Driven Behavior Evaluation and Assurance in Foundation-Model Systems for Mental Health Screening

TL;DR

CPEMH is a modular agentic framework that systematically evaluates and assures prompt-driven behavior in foundation models for mental health screening, emphasizing stability, traceability, and robustness.

Contribution

It introduces an orchestrated, modular architecture for behavioral assurance in large-scale language systems applied to mental health screening tasks.

Findings

Demonstrated capacity to stabilize and audit model behavior in depression screening.

Highlighted importance of modular orchestration for behavioral assurance.

Emphasized stability over architectural complexity in system design.

Abstract

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.