Loading...
Generative recommendation of customized instructional activities for virtual laboratories
Hasan, Touseef Binte Kader, Faria Dey, Indrani Narayanan, Hari Puntambekar, Sadhana Karmaker, Santu
Hasan, Touseef
Binte Kader, Faria
Dey, Indrani
Narayanan, Hari
Puntambekar, Sadhana
Karmaker, Santu
Citations
Altmetric:
Files
Loading...
Hasan_2026.pdf
Adobe PDF, 58.39 KB
Other Names
Location
Time Period
Advisors
Original Date
Digitization Date
Issue Date
2026-03-24
Type
Abstract
Genre
Keywords
Subjects (LCSH)
Citation
Abstract
Generating instructional activities that align with the constraints and objectives of virtual laboratories is labor-intensive and difficult to scale. To address this challenge, we propose the Generative Recommendation Framework (GRF) for the constrained generation of instructional activities. GRF guides Large Language Models (LLMs) to generate instructionally aligned activities under varying levels of instructional control using a four-level prompt design based on the TELeR taxonomy of prompting. We evaluate our framework across 27 LLMs on five virtual labs, producing 5,400 activities. Using a five-criterion rubric with both LLM-as-a-judge and human evaluation, we observe a notable alignment between human and LLM-based judgments. Our experiments show that instruction-tuning enables open-source LLMs to generate high-quality structured activities compared to their base variants. Moreover, criterion-wise analysis reveals that current LLMs consistently struggle to produce activities with clear structure and step-wise conceptual progression. These findings help characterize how LLMs currently perform on constrained activity generation for virtual labs and pinpoint where further improvements are needed to provide real-time guidance to students.
Table of Contents
Description
Poster project completed at Wichita State University, School of Computing; University of Central Florida, Department of Computer Science; University of Wisconsin-Madison, Department of Educational Psychology; and Department of Computer Science and Software Engineering, Auburn University.
Presented at the 23rd Annual Capitol Graduate Research Summit, Topeka, KS, March 24, 2026.
Presented at the 23rd Annual Capitol Graduate Research Summit, Topeka, KS, March 24, 2026.
Publisher
Wichita State University