A Human-in-the-Loop Framework for AI-Assisted Scoring in Large-Scale Writing Assessment

Human-in-the-loop strategies are being integrated into healthcare and educational assessments to maintain quality. In healthcare, the framework prioritizes clinician command based on risk, reversibility, and stakes, ensuring AI recommends while humans remain accountable for patient-touching decisions. Similarly, in large-scale writing assessments, AI and LLMs are utilized to enhance grading efficiency and scalability for short texts. This approach reduces manual workloads while preserving assessment quality through a validated human-in-the-loop strategy within real operational contexts.