Search papers, labs, and topics across Lattice.
This paper analyzes the challenges of evaluating LLM-based software engineering (AI4SE) tools, highlighting the limitations of traditional evaluation metrics due to the open-ended, non-deterministic nature of LLM outputs. It surveys current evaluation practices in AI4SE, revealing their shortcomings in realistic settings, such as the lack of stable ground truth and the subjectivity of quality assessment. The paper concludes by outlining future research directions for developing more robust and scalable evaluation methodologies tailored to the unique characteristics of LLMs in software engineering.
Evaluating LLM-powered software engineering tools is fundamentally broken, as traditional metrics fail to capture the nuanced, non-deterministic nature of their outputs.
Large Language Models (LLMs) are increasingly embedded in software engineering (SE) tools, powering applications such as code generation, automated code review, and bug triage. As these LLM-based AI for Software Engineering (AI4SE) systems transition from experimental prototypes to widely deployed tools, the question of what it means to evaluate their behavior reliably has become both critical and unanswered. Unlike traditional SE or machine learning systems, LLM-based tools often produce open-ended, natural language outputs, admit multiple valid answers, and exhibit non-deterministic behavior across runs. These characteristics fundamentally challenge long-standing evaluation assumptions such as the existence of a single ground truth, deterministic outputs, and objective correctness. In this paper, we examine LLM evaluation as a general, task-dependent concept through the lens of SE tasks. We discuss why reliable evaluation is essential for trust, adoption, and meaningful assessment of LLM-based tools, summarize the current state of evaluation practices, and highlight their limitations in realistic AI4SE settings. We then identify key challenges facing current approaches, including the absence of stable ground truth, subjectivity and multi-dimensional quality, evaluation instability due to non-determinism, limitations of automated and model-based evaluation, and fragmentation of evaluation practices. Finally, we outline future directions aimed at advancing LLM evaluation toward more robust, scalable, and trustworthy methodologies, to stimulate discussion on principled evaluation practices that can keep pace with the growing role of LLMs in SE.