LLM4AlgorithmDesign

Collection on Algorithm Design with Large Language Models.

🔥 Applying Large language models (LLMs) for algorithm design (AD) is an emerging research area. This is a collection of references and papers of LLM4AD (with focus on optimization algorithms). The Papers are sorted by time (first publicly available).

For more details, please see our survey paper:

• A Systematic Survey on Large Language Models for Algorithm Design (ACM Computing Surveys, 2025)

@article{liu2025systematic,
  author = {Liu, Fei and Yao, Yiming and Guo, Ping and Yang, Zhiyuan and Lin, Xi and Zhao, Zhe and Tong, Xialiang and Mao, Kun and Lu, Zhichao and Wang, Zhenkun and Yuan, Mingxuan and Zhang, Qingfu},
  title = {A Systematic Survey on Large Language Models for Algorithm Design},
  year = {2025},
  journal = {ACM Computing Surveys}
}

Video Introductions and Slides:

• English Webinar (IEEE CIS)
• Chinese Introduction (中文视频介绍)

Any suggestions and pull requests are welcomed!

It is far from a comprehensive list. If you want to update the list:

• Fork, Add, and Merge
• Report an issue
• Contact Fei Liu (fliu36-c@my.cityu.edu.hk)

The sharing principle of these references here is for research. If any authors do not want their paper to be listed here, please feel free to contact us.

Overview

• Platform
• Course
• Tutorial
• Competition
• Special Issues
• Research Papers in Four Paradigms
• Research Papers
  • Review
  • Position Paper
  • Algorithm/Heuristic/Function Search
  • LLM as optimizer
  • Code Generation
  • Prompt Opt.
  • Machine Learning
  • Science
  • Industry
• Related Collections

Platform

Project	Description
LLM4AD	Open-source Python-based Platform leveraging Large Language Models (LLMs) for Automatic Algorithm Design (AD) with 100+ tasks and 10+ methods
BLADE	Benchmarking LLM-driven Automated Design and Evolution of Iterative Optimization Heuristics
EASE	Effortless Algorithmic Solution Evolution is a framework that leverages Large Language Models (LLMs) to generate solutions (algorithms, text, images, etc.) based on user-defined parameters. It provides a flexible and adaptive approach to automated problem-solving.

Course

Course	Description
2024 Fall, LLM Agents	LLM basics and LLM for agents

Tutorial&Workshop

|