Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Efficient and Green Large Language Models for Software Engineering: Literature Review, Vision, and the Road Ahead

Jieke Shi, Zhou Yang, David Lo

TL;DR

This survey addresses the need for efficient and green LLM-based softwareengineering tools (LLM4SE) by defining a six-dimensional framework for efficiency and sustainability, surveying data-, model-, system-, and program-centric approaches, and outlining concrete roadmap steps. It highlights techniques such as PEFT, model compression, architecture upgrades, dynamic inference, and grammar-based program optimization, with quantified gains where available. The paper argues that integrating efficiency and greenness yields the most practical impact, enabling broader access for individual practitioners and SMEs while lowering environmental footprints. By 2030, the vision is low-cost, private, and green LLM4SE tools embedded in local workflows that accelerate software development and promote societal sustainability.

Abstract

Large Language Models (LLMs) have recently shown remarkable capabilities in various software engineering tasks, spurring the rapid growth of the Large Language Models for Software Engineering (LLM4SE) area. However, limited attention has been paid to developing efficient LLM4SE techniques that demand minimal computational cost, time, and memory resources, as well as green LLM4SE solutions that reduce energy consumption, water usage, and carbon emissions. This paper aims to redirect the focus of the research community towards the efficiency and greenness of LLM4SE, while also sharing potential research directions to achieve this goal. It commences with a brief overview of the significance of LLM4SE and highlights the need for efficient and green LLM4SE solutions. Subsequently, the paper presents a vision for a future where efficient and green LLM4SE revolutionizes the LLM-based software engineering tool landscape, benefiting various stakeholders, including industry, individual practitioners, and society. The paper then delineates a roadmap for future research, outlining specific research paths and potential solutions for the research community to pursue. While not intended to be a definitive guide, the paper aims to inspire further progress, with the ultimate goal of establishing efficient and green LLM4SE as a central element in the future of software engineering.

Efficient and Green Large Language Models for Software Engineering: Literature Review, Vision, and the Road Ahead

TL;DR

This survey addresses the need for efficient and green LLM-based softwareengineering tools (LLM4SE) by defining a six-dimensional framework for efficiency and sustainability, surveying data-, model-, system-, and program-centric approaches, and outlining concrete roadmap steps. It highlights techniques such as PEFT, model compression, architecture upgrades, dynamic inference, and grammar-based program optimization, with quantified gains where available. The paper argues that integrating efficiency and greenness yields the most practical impact, enabling broader access for individual practitioners and SMEs while lowering environmental footprints. By 2030, the vision is low-cost, private, and green LLM4SE tools embedded in local workflows that accelerate software development and promote societal sustainability.

Abstract

Large Language Models (LLMs) have recently shown remarkable capabilities in various software engineering tasks, spurring the rapid growth of the Large Language Models for Software Engineering (LLM4SE) area. However, limited attention has been paid to developing efficient LLM4SE techniques that demand minimal computational cost, time, and memory resources, as well as green LLM4SE solutions that reduce energy consumption, water usage, and carbon emissions. This paper aims to redirect the focus of the research community towards the efficiency and greenness of LLM4SE, while also sharing potential research directions to achieve this goal. It commences with a brief overview of the significance of LLM4SE and highlights the need for efficient and green LLM4SE solutions. Subsequently, the paper presents a vision for a future where efficient and green LLM4SE revolutionizes the LLM-based software engineering tool landscape, benefiting various stakeholders, including industry, individual practitioners, and society. The paper then delineates a roadmap for future research, outlining specific research paths and potential solutions for the research community to pursue. While not intended to be a definitive guide, the paper aims to inspire further progress, with the ultimate goal of establishing efficient and green LLM4SE as a central element in the future of software engineering.
Paper Structure (32 sections, 3 figures)

This paper contains 32 sections, 3 figures.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Large Language Models (LLMs)
  4. Encoder-Only LLMs.
  5. Encoder-Decoder LLMs.
  6. Decoder-Only LLMs.
  7. Definition of Efficient and Green LLM4SE
  8. Significance of Efficient and Green LLM4SE
  9. Why Efficient LLM4SE?
  10. Why Green LLM4SE?
  11. Why Synergizing Efficient and Green LLM4SE?
  12. Literature Review
  13. Data-Centric Efficient and Green LLM4SE
  14. Model-Centric Efficient and Green LLM4SE
  15. Model Compression.
  16. ...and 17 more sections

Figures (3)

  • Figure 1: Six Dimensions of Efficient and Green LLM4SE, along with their metrics.
  • Figure 2: Current Techniques for Efficient and Green LLM4SE
  • Figure 3: Roadmap for Efficient and Green LLM4SE