ARCHED: A Human-Centered Framework for Transparent, Responsible, and Collaborative AI-Assisted Instructional Design

Mon, 31 Mar 2025 00:00:00 +0000

How can LLMs assist instructional design while preserving educator control and pedagogical alignment? ARCHED separates learning-objective generation from analysis, combining a Learning Objective Generation System (LOGS) with an Objective Analysis Engine (OAE). Educators specify the teaching context, inspect Bloom’s taxonomy classifications, and select or refine objectives before assessment development.

The published evaluation includes expert classification of 120 computer-science learning objectives, a comparison of 30 generated and 30 human-created objectives, and preliminary feedback from two educational experts. It examines objective quality, taxonomic alignment, and the practical workflow. This evidence concerns instructional-design support; it does not establish classroom learning gains or broad adoption outcomes. The framework is relevant to research on human-AI collaboration, transparent generation and evaluation workflows, and responsible AI in education.

The workshop took place March 3, 2025; the was published March 31, 2025.

Learning Objectives | Yizirui Fang

ARCHED: A Human-Centered Framework for Transparent, Responsible, and Collaborative AI-Assisted Instructional Design