Designing AI-Resilient Assessments Using Interconnected Problems: A Theoretically Grounded and Empirically Validated Framework

📝 Original Info

• Title: Designing AI-Resilient Assessments Using Interconnected Problems: A Theoretically Grounded and Empirically Validated Framework
• ArXiv ID: 2512.10758
• Date: 2025-12-11
• Authors: Kaihua Ding

📝 Abstract

The proliferation of generative AI tools has rendered traditional modular assessments in computing and datacentric education obsolete, creating a critical disconnect between academic training and industry practice. This paper presents a theoretically grounded framework for designing AI-resilient assessments, supported by proofs and empirical validation. We make three primary contributions. First, we establish two formal propositions: (1) assessments composed of interconnected problems, where outputs serve as inputs to subsequent stages, are inherently more AI-resilient than modular assessments due to multi-step reasoning and context limitations of large language models; and (2) semi-structured problems with deterministic success criteria provide more reliable measures of student competency than fully open-ended projects, which allow AI systems to default to familiar solution patterns. These findings challenge widely cited recommendations-discussed in a UNESCO-hosted analysis [1] and institutional guidelines [2], [3], [4] -that open-ended assessments maintain academic integrity and encourage deeper material engagement in the AI era-an assumption our findings contradict. Second, we validate these propositions throug...

📄 Full Content

...(본문 내용이 길어 생략되었습니다. 사이트에서 전문을 확인해 주세요.)