Wouldn't every single participant need a copy of the entire training set? | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		LeoPanthera on May 13, 2024 \| parent \| context \| favorite \| on: Release of Fugaku-LLM – a large language model tra... Wouldn't every single participant need a copy of the entire training set?

samus on May 14, 2024 [–]

That's the next big problem. And there need to be mechanisms to ensure that the network is not poisoned with undesirable input.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact