Researchers propose a research roadmap for merging search-based software engineering with AI foundation models across three directions
Synopsis
Presented as a research roadmap, the work surveys the current relationship between search-based software engineering (SBSE), a field active for about 25 years, and AI foundation models (FMs) such as large language models, analyzing three core aspects—using FMs to enhance SBSE, applying SBSE to advance FMs, and exploring their integration—while identifying open challenges and potential research directions and envisioning the future of SBSE in the era of FMs.
Figure 1: Roadmap overview showing the discussion flow: strengths and weaknesses ( Section 3 ), SBSE–FM synergies ( Sections 4 , 5 and 6 ), empirical evaluation challenges ( Section 7 ), and the 2030 research horizon ( Section 8 ).
arXivInterpretation
The article proposes and organizes a research roadmap that characterizes the current landscape of SBSE in relation to foundation models (FMs). SBSE already has roughly 25 years of accumulated work spanning the entire software engineering lifecycle, yet how it evolves alongside FMs remains undetermined; the article turns this window of uncertainty into an explicit roadmap agenda. This is a position and survey-style argument grounded in an overall judgment about SBSE history and the rise of FMs, not in experimental data or system benchmarks.
The article decomposes the SBSE–FM relationship into three core aspects: utilizing FMs to enhance SBSE, applying SBSE to advance FMs, and exploring the integration of SBSE and FMs. It refines the broad topic of combining AI with software engineering into three strands, including a bidirectional and integrative one, offering a classification frame for later work. The three-aspect division is the article's own stated analytical structure, a conceptual framework without quantitative validation in the text.
The article identifies open challenges and outlines potential research directions, adding a forward-thinking perspective that highlights research opportunities for addressing challenges in emerging domains. It goes beyond describing the landscape to convert it into an actionable research agenda aimed at emerging-domain challenges. This is an opinion-based outlook whose value lies in agenda setting rather than completed empirical findings.
Perspective
The article is positioned to provide a research agenda for the SBSE–FM intersection: it suits researchers choosing topics in this intersection and software engineering practitioners who need to understand where SBSE stands in the era of FMs, helping them form question lists and directional judgments. Its conclusions apply to discussions that treat SBSE and FMs as complementary and potentially integrable, not to evaluating the performance of a specific tool or method.
Because the loaded text is summary-level, the specific challenge items, example directions, and any supporting cases under the three aspects are not presented, so the maturity and priority of each direction cannot be judged. Readers should still watch which of these directions already have reproducible empirical support and which remain proposals, and how the integration strand would connect with established SBSE workflows in practice.