Four Types of LLM Reliance and Their Predictors Among Undergraduate Writers: A Mixed-Methods Study at a Minority-Serving R1 University
arXiv:2606.28749v1 Announce Type: cross Abstract: Although most undergraduates now use large language models (LLMs), a form of generative artificial intelligence (GenAI) for academic writing, no validated method distinguishes the qualitatively different ways students rely on them. Existing instruments assess reliance solely by frequency of use, a measure that, as this study shows, inadvertently rewards dependence on AI rather than recognizing students' own intellectual contribution. Conducted at a public minority-serving university and grounded in the AI Literacy Framework, Expectancy-Value Th
The proliferation of LLMs in academic settings necessitates understanding their actual impact on student learning and output beyond mere frequency of use.
This study challenges simplistic metrics of LLM reliance, highlighting the need for nuanced assessment methods that differentiate between constructive tool use and over-dependence, which is crucial for educational policy and AI integration.
The understanding of how students interact with LLMs moves beyond simple usage rates to a more complex typology of reliance, impacting how educational institutions might structure guidelines and assessments.
- · AI literacy frameworks developers
- · Educational researchers
- · Students with high intellectual contribution
- · Developers of simplistic AI detection tools
- · Institutions relying solely on frequency metrics
- · Students who over-rely on LLMs for intellectual output
Educational institutions will seek more sophisticated methods to evaluate AI usage in academic work.
New pedagogical approaches and evaluation tools will emerge to foster beneficial LLM integration while discouraging undue reliance.
The development of LLMs themselves may be influenced by these findings, pushing towards features that encourage critical thinking rather than passive generation.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI