LogicCraft: LLM-Assisted Optimization of Netlist to Layout for Complex Custom Standard Cell Designs

  • Minwei Lin
  • , Duy Hieu Bui
  • , Chao Wang
  • , Wangzilu Lu
  • , Ruoyu Tang
  • , Qing Zhang
  • , Yuhang Zhang
  • , Jian Zhao
  • , Xuan Tu Tran*
  • , Yongfu Li*
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In digital integrated circuit (IC) design, the need for custom standard cells tailored to specific design constraints, such as power, area, and performance, is increasingly critical. This paper presents an approach that leverages Large Language Models (LLMs) to automate the translation and optimization of Boolean tables into custom pull-up networks, using established principles to generate complementary pull-down networks. The proposed methodology focuses on automated translation, optimization, and seamless integration into standard electronic design automation (EDA) workflows. By utilizing LLMs, this approach significantly reduces design time, enhances optimization, and minimizes human error. This makes it a powerful tool for improving the efficiency and effectiveness of custom standard cell design in digital logic circuits.

Original languageEnglish
Title of host publicationAICAS 2025 - 2025 7th IEEE International Conference on Artificial Intelligence Circuits and Systems, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331524241
DOIs
StatePublished - 2025
Externally publishedYes
Event7th IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2025 - Bordeaux, France
Duration: 28 Apr 202530 Apr 2025

Publication series

NameAICAS 2025 - 2025 7th IEEE International Conference on Artificial Intelligence Circuits and Systems, Proceedings

Conference

Conference7th IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2025
Country/TerritoryFrance
CityBordeaux
Period28/04/2530/04/25

Keywords

  • Large Language Model
  • Logic Expression
  • Optimization
  • SPICE
  • Standard Cell Design

Fingerprint

Dive into the research topics of 'LogicCraft: LLM-Assisted Optimization of Netlist to Layout for Complex Custom Standard Cell Designs'. Together they form a unique fingerprint.

Cite this