NeurIPS2024 Edge-Devices LLM Challenge
Shiwei Liu
🚀 Join the NeurIPS Edge-Devices LLM Challenge! 🚀 #NeurIPS2024
Are you ready to push the boundaries of Large Language Models (LLMs) on edge devices? Participate in our NeurIPS challenge and showcase your skills in optimizing LLMs for real-world applications! With a total prize pool of $40,000, this is your chance to shine.
Discord: https://discord.gg/yD89SPNr3b
Register: https://docs.google.com/forms/d/e/1FAIpQLSfDfgHqbEdT479jK5u2ECWNxix5wc9o8vcUQ7KoVWehp4K8vg/viewform
Challenge Overview
Our challenge focuses on making LLMs more efficient and effective for deployment on edge devices. We offer two exciting tracks:
- Compression Challenge: Demonstrate your ability to compress LLMs without compromising performance. Show how you can reduce the model size and memory footprint while maintaining or even improving accuracy.
- Training from Scratch Challenge: Prove your expertise by training an LLM from scratch, specifically tailored for edge devices. Optimize the training process to create a robust and efficient model.
Why Participate?
- Win Big: Compete for a share of the $40,000 prize pool.
- Innovate: Contribute to cutting-edge research in LLM optimization.
- Network: Connect with leading experts and peers in the field.
- Showcase: Gain recognition for your work at one of the premier AI conferences.
Key Dates
- Registration Opens: June 25th, 2024
- Registration Ends: July 25th, 2024
- Submission Deadline: October 25th, 2024
- Winners Announced: November 20th, 2024
How to Join
- Register for the challenge on the competition website via Google Forms.
- Choose your track(s): Compression Challenge, Training from Scratch Challenge, or both!
- Submit your solutions by the deadline.
Evaluation
Submissions will be evaluated on a diverse set of tasks, memory usage, and throughput. Detailed evaluation criteria will be provided on the challenge page.
Ready to Accept the Challenge?
Don't miss this opportunity to make a significant impact in the field of LLMs and edge devices. Register now and start preparing your submissions!
Let's advance the future of LLMs together. Good luck, and see you at NeurIPS 2024! 🌟
Shiwei Liu, Ph.D.
Newton International Fellow,
University of Oxford, UK