from
Hacker News
Top
New
It's remarkably easy to inject new medical misinformation into LLMs
by
prng2021
on 1/10/25, 2:45 AM with 1 comments
by
gnabgib
on 1/10/25, 3:43 AM
The paper: [Medical large language models are vulnerable to data-poisoning attacks](
https://news.ycombinator.com/item?id=42640260
)