Make Each Iteration Count

被引：0

作者：

Hu, Yuchen ^{[1
]}

Xu, Ke ^{[1
]}

Sun, Jialin ^{[1
]}

Fang, Xinwei ^{[2
]}

Shan, WeiWei ^{[1
]}

Wang, Xi ^{[1
]}

Jiang, Zhe ^{[1
]}

机构：

[1] Southeast Univ, Natl Ctr Technol Innovat EDA, Sch Integrated Circuits, Nanjing, Peoples R China

[2] Univ York, Dept Comp Sci, York, England

来源：

PROCEEDINGS OF THE ACM TURING AWARD CELEBRATION CONFERENCE-CHINA 2024, ACM-TURC 2024 | 2024年

关键词：

D O I：

10.1145/3674399.3674482

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Large Language Models (LLMs) is widely used for code debugging (e.g., C and Python) but face limitations in debugging Register Transfer Level (RTL) code largely due to data scarcity. This paper presents Make Each Iteration Count (MEIC), a novel framework for RTL debugging. Unlike traditional approaches heavily relying on prompt engineering, model tuning, and model training, MEIC employs an iterative process with LLM to address syntax and function errors efficiently. We also introduce an open-source dataset with 178 RTL errors for evaluation. Results demonstrate a 93% fix rate for syntax errors and a 78% fix rate for function errors.

引用

页码：236 / 238

页数：3

共 7 条

[1]

DeLorenzo M, 2024, Arxiv, DOI arXiv:2402.03289

[2] RTL-REPAIR: Fast Symbolic Repair of Hardware Design Code [J].

Laeufer, Kevin ;

Fajardo, Brandon ;

Ahuja, Abhik ;

Iyer, Vighnesh ;

Nikolic, Borivoje ;

Sen, Koushik .

PROCEEDINGS OF THE 29TH ACM INTERNATIONAL CONFERENCE ON ARCHITECTURAL SUPPORT FOR PROGRAMMING LANGUAGES AND OPERATING SYSTEMS, VOL 3, ASPLOS 2024, 2024, :867-881

[3] Invited Paper: VerilogEval: Evaluating Large Language Models for Verilog Code Generation [J].

Liu, Mingjie ;

Pinckney, Nathaniel ;

Khailany, Brucek ;

Ren, Haoxing .

2023 IEEE/ACM INTERNATIONAL CONFERENCE ON COMPUTER AIDED DESIGN, ICCAD, 2023,

[4]

Tsai YD, 2024, Arxiv, DOI arXiv:2311.16543

[5]

Wei Ran, 2023, IEEE Transactions on ComputerAided Design of Integrated Circuits and Systems

[6]

Yao XF, 2024, Arxiv, DOI arXiv:2403.11671

[7]

Zhang Y, 2023, Arxiv, DOI [arXiv:2309.01219, 10.48550/arXiv.2309.01219, DOI 10.1162/COLI.A.16, DOI 10.48550/ARXIV.2309.01219]

← 1 →