Reproduce “Beyond Static Scoring: Enhancing Assessment Validity via AI-Generated Interactive Verification” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Tom Lee, Sihoon Lee, Seonghun Kim
This research introduces a two-stage Human-AI Collaboration framework that integrates automated rubric generation and scoring with AI-generated interactive verification questions to enhance the fairness and validity of open-ended assessments. A pilot study involving university instructors found high usability and perceived effectiveness in revealing students' deeper understanding, with unanimous agreement on its positive contribution to assessment fairness and validity.
Key insights
- The framework shifts the assessment paradigm from unreliable AI detection to a robust, AI-augmented verification of genuine student understanding, regardless of initial content generation methods.
- It demonstrates that procedural fairness, achieved through consistent automated rubric-based scoring, and construct validity, enhanced by interactive, evidence-based verification, can be synergistically combined within a single system.
- AI is effectively positioned as an intelligent assistant to automate mundane assessment tasks and generate insightful prompts, thereby augmenting human educators' capabilities and allowing them to focus on expert judgment.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.