News for hackers. Stuff that matters.
Re:Memorizing Transformers by visarga on Friday May 20, 2022 @04:17PM (#31452026)
> No concern that this enables β€˜Tay’ like failure modes where a system can be manipulated through input into generating particular output?

Isn't that the core idea in prompting and few shot learning for large language models?

There are no comments on this comment yet.