Large language models are … large. Forget Bitcoin and recharging electric vehicles; the grid could be toppled by powering AI in a few years. It would be optimal if AI could run on more underpowered edge devices. What if there was a quantum-inspired way to make LLMs smaller without sacrificing overall performance in combined metrics? We explore a way to do that and other advanced ideas, like selectively removing information from models, in this episode. Join Host Konstantinos Karagiannis for a chat about how businesses can solve many of the woes associated with bringing AI in-house with Sam Mugel from Multiverse.
For more on Multiverse, visit https://multiversecomputing.com/.
Read the CompactifAI paper: https://arxiv.org/abs/2401.14109.
Visit Protiviti at www.protiviti.com/US-en/technology-consulting/quantum-computing-services to learn more about how Protiviti is helping organizations get post-quantum ready.
Follow host Konstantinos Karagiannis on all socials: @KonstantHacker and follow Protiviti Technology on LinkedIn and Twitter: @ProtivitiTech.
Questions and comments are welcome!
Theme song by David Schwartz, copyright 2021.
The views expressed by the participants of this program are their own and do not represent the views of, nor are they endorsed by, Protiviti Inc., The Post-Quantum World, or their respective officers, directors, employees, agents, representatives, shareholders, or subsidiaries. None of the content should be considered investment advice, as an offer or solicitation of an offer to buy or sell, or as an endorsement of any company, security, fund, or other securities or non-securities offering. Thanks for listening to this podcast. Protiviti Inc. is an equal opportunity employer, including minorities, females, people with disabilities, and veterans.

Quantum AI Compression: Smaller LLMs without Losing Performance — with Sam Mugel of Multiverse

The Post-Quantum World