2026 ASEE Annual Conference & Exposition

Systematic Literature Reviews (SLR) are essential for synthesizing scientific knowledge. One of the challenges of SLRs is that they require a lot of time for repetitive tasks such as data collection and the screening process based on inclusion/exclusion criteria aligned with the scope of the research question(s). Recent advances in Machine Learning (ML) and Large Language Models (LLMs) offer new opportunities to support researchers throughout the SLR process. Due to the multifaceted nature of SLRs, which deal with large-scale and heterogeneous data, computational tools help support and manage this complexity while revealing patterns that might not be visible through manual analysis. In this study, ML and LLMs were integrated across the main stages of the review to refine searches, support screening, data collection, and synthesis. The use of ML enables the classification, organization, and visualization of a large number of studies. At the same time, the use of LLMs supports the understanding, comparison, and synthesis of textual data. Together, these tools show how researchers and computation can jointly strengthen and facilitate research processes such as SLRs. This Work in Progress explores the integration of computer-assisted methods into the development of SLRs, following the PRISMA framework’s four phases and 27 items. We explore how ML and LLM-based tools can automate tasks such as literature search, screening, data collection, and synthesis while preserving methodological trustworthiness, validity, and replicability through human verification and cyclical processes. Our study highlights emerging practices and discusses ethical considerations at each stage, including transparency, energy consumption, potential decrease of researchers’ skills, and data privacy. We argue that computational tools do not replace human expertise, instead, they can serve to strengthen the scalability of analysis guided by methods, human verification, and cyclical processes. The findings aim to offer a set of good practices for the responsible, ethical, and effective use of computer- and AI-assisted methods in SLRs.