Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
SirMaster
on April 11, 2024
|
parent
|
context
|
favorite
| on:
Anyone got a contact at OpenAI. They have a spider...
I thought LLMs don't say when they don't know something because of how they are tuned and because of RLHF.
ben_w
on April 11, 2024
[–]
They can say they don't know, and have been trained to in at least some cases; I think the deeper problem — which we don't know how to fix in humans, the closest we have is the scientific method — is they can be confidently wrong.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
