Software Engineering for Large Language Models: Research Status, Challenges and the Road Ahead

TL;DR

This paper systematically reviews the current state, challenges, and future research directions in applying software engineering principles to the development lifecycle of large language models, aiming to guide future advancements.

Contribution

It provides a comprehensive analysis of LLM development phases from an SE perspective and identifies key challenges and research opportunities for each phase.

Findings

Identifies six phases in LLM development lifecycle.

Highlights key challenges in each development phase.

Suggests potential research directions for improving LLM development.

Abstract

The rapid advancement of large language models (LLMs) has redefined artificial intelligence (AI), pushing the boundaries of AI research and enabling unbounded possibilities for both academia and the industry. However, LLM development faces increasingly complex challenges throughout its lifecycle, yet no existing research systematically explores these challenges and solutions from the perspective of software engineering (SE) approaches. To fill the gap, we systematically analyze research status throughout the LLM development lifecycle, divided into six phases: requirements engineering, dataset construction, model development and enhancement, testing and evaluation, deployment and operations, and maintenance and evolution. We then conclude by identifying the key challenges for each phase and presenting potential research directions to address these challenges. In general, we provide valuable insights from an SE perspective to facilitate future advances in LLM development.