SLRMentor uses LLMs to teach systematic literature review process

SLRMentor is a new conversational assistant powered by large language models, designed to support both learning about the systematic literature review (SLR) process and executing planning activities in software engineering. Developed by researchers from multiple institutions, the tool offers general guidance on SLR methodology and assists with critical planning tasks such as constructing effective search strings and reasoning about inclusion and exclusion criteria. All explanations are grounded in established SLR guidelines, ensuring that users receive accurate, standardized advice. The tool functions as an interactive tutor, allowing novices to ask questions and receive step-by-step guidance without replacing the need for active methodological judgment.

In a pilot validation with graduate students, SLRMentor demonstrated significant benefits: it helped clarify the SLR process and planning decisions, lowered the initial barriers that often discourage novice researchers, and supported learning while still requiring users to apply their own judgment. The tool does not automate the entire SLR — instead, it acts as a knowledgeable assistant that explains concepts and helps users think through decisions. This approach balances the power of LLMs with the pedagogical goal of teaching rigorous research methodologies. The paper (arXiv:2606.07831) provides full details and suggests that SLRMentor could be a valuable addition to software engineering research methods courses.