Home > Technology peripherals > AI > body text

You are sincerely invited to participate in the LLM4HWDesign@ICCAD2024 competition jointly organized by Georgia Tech and Nvidia!

WBOY
Release: 2024-08-07 18:22:53
Original
1011 people have browsed it

The LLM4HWDesign@ICCAD 2024 competition jointly organized by Georgia Tech EIC Laboratory and Nvidia Corporation has officially launched! This competition aims to promote the performance of large language models (LLM) in auxiliary hardware design. Elites from all walks of life are sincerely invited to actively participate and jointly explore methods of automated data generation, collection, cleaning and annotation, and build an open source, large-scale, high-quality The hardware code data set realizes revolutionary changes in the field of LLM-assisted hardware design.

You are sincerely invited to participate in the LLM4HWDesign@ICCAD2024 competition jointly organized by Georgia Tech and Nvidia!

Competition Goals

The main goals of the competition are:

  • To improve the performance and application effects of LLM in the field of auxiliary hardware design by exploring methods of constructing and labeling data sets.
  • Solve problems such as data collection, generation, filtering and annotation to achieve efficient hardware design automation.
  • Improve the performance of LLM in the field of auxiliary hardware design and build open source, high-quality hardware code data sets.

Contest Process

Phase 1 (July 7th to August 10th): Data sample collection and generation

  • Collect and generate data samples (Verilog code) related to hardware design.
  • Data collection: Collect new samples from open source code bases, academic papers and proprietary designs, ensuring open source and public use.
  • Data generation: Use LLM or other tools to generate new samples, ensuring open source and public use.

Second Phase (August 20th to October 1st): Data Screening and Annotation

  • Screen and annotate the data collected and generated in the first phase.
  • Data filtering: Develop automated techniques to remove low-quality samples.
  • Accurate Description: Generate more accurate descriptions for samples.
  • Label Design: Create labeling strategies to close the knowledge gap during LLM pre-training and fine-tuning.

    You are sincerely invited to participate in the LLM4HWDesign@ICCAD2024 competition jointly organized by Georgia Tech and Nvidia!

    Award setting

? First place: 1 RTX 4080 GPU per team + $2000
? Second place: 1 RTX 4080 GPU per team + $1000
? Third place: 1 RTX 4070 GPU per team + $500

Registration and more information

Registration deadline is July 30th. If contestants are willing to:

● Participate in cutting-edge research on hardware design based on large language models
● Receive awards on-site at ICCAD and collaborate on publishing related papers
● Compete for rich prizes

Please visit the competition webpage for more details and registration information:

  • [LLM4HWDesign Competition Home Page](https://nvlabs.github.io/LLM4HWDesign/)
  • [Google Form Registration Link (Recommended)](https://forms.gle/aUxfXSkeQhctQEdF6)
  • [Questionnaire Star Registration Link (alternative)](https://www.wjx.cn/vm/O0ZQUao.aspx)

If you have any other questions, please contact llm4hwdesign@groups.gatech.edu.

We look forward to your participation and work with us to promote innovation and development in the field of hardware design!

The above is the detailed content of You are sincerely invited to participate in the LLM4HWDesign@ICCAD2024 competition jointly organized by Georgia Tech and Nvidia!. For more information, please follow other related articles on the PHP Chinese website!

Related labels:
source:jiqizhixin.com
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Popular Tutorials
More>
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template