A Multi-Expert Large Language Model Architecture for Verilog Code Generation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Nadimi, Bardia, Zheng, Hao
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914965051408384
author Nadimi, Bardia
Zheng, Hao
author_facet Nadimi, Bardia
Zheng, Hao
contents Recently, there has been a surging interest in using large language models (LLMs) for Verilog code generation. However, the existing approaches are limited in terms of the quality of the generated Verilog code. To address such limitations, this paper introduces an innovative multi-expert LLM architecture for Verilog code generation (MEV-LLM). Our architecture uniquely integrates multiple LLMs, each specifically fine-tuned with a dataset that is categorized with respect to a distinct level of design complexity. It allows more targeted learning, directly addressing the nuances of generating Verilog code for each category. Empirical evidence from experiments highlights notable improvements in terms of the percentage of generated Verilog outputs that are syntactically and functionally correct. These findings underscore the efficacy of our approach, promising a forward leap in the field of automated hardware design through machine learning.
format Preprint
id arxiv_https___arxiv_org_abs_2404_08029
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Multi-Expert Large Language Model Architecture for Verilog Code Generation
Nadimi, Bardia
Zheng, Hao
Machine Learning
Artificial Intelligence
Programming Languages
Software Engineering
Recently, there has been a surging interest in using large language models (LLMs) for Verilog code generation. However, the existing approaches are limited in terms of the quality of the generated Verilog code. To address such limitations, this paper introduces an innovative multi-expert LLM architecture for Verilog code generation (MEV-LLM). Our architecture uniquely integrates multiple LLMs, each specifically fine-tuned with a dataset that is categorized with respect to a distinct level of design complexity. It allows more targeted learning, directly addressing the nuances of generating Verilog code for each category. Empirical evidence from experiments highlights notable improvements in terms of the percentage of generated Verilog outputs that are syntactically and functionally correct. These findings underscore the efficacy of our approach, promising a forward leap in the field of automated hardware design through machine learning.
title A Multi-Expert Large Language Model Architecture for Verilog Code Generation
topic Machine Learning
Artificial Intelligence
Programming Languages
Software Engineering
url https://arxiv.org/abs/2404.08029