Hacker News

2171 readers

2 users here now

A mirror of Hacker News' best submissions.

founded 1 year ago

MODERATORS

[email protected]

Resume Tip: Hacking "AI" screening of resumes (www.solipsys.co.uk)

submitted 6 months ago by [email protected] to c/[email protected]

9 comments fedilink hide all child comments

HN Discussion

you are viewing a single comment's thread
view the rest of the comments

[–] [email protected] 7 points 6 months ago (3 children)

I don't think the models absorb new information from questions, do they?

[–] [email protected] 6 points 6 months ago

I'd be surprised if they didn't, to be honest.

[–] [email protected] 5 points 6 months ago (1 children)

They do not. My understanding is that the models are static.

[–] Infinite 4 points 6 months ago (1 children)

Their settings imply that they do

[–] [email protected] 7 points 6 months ago* (last edited 6 months ago)

I guess I should be a bit more specific. There's no feedback loop into the currently running model. They could potentially use that data to tune or train future versions of the model, but there's no mechanism to live update the model weights in place. Or at least not to my knowledge.

The only things that'll make a difference in the same version of the model are the system prompt, context window and prompt.

[–] [email protected] 1 points 6 months ago

Yeah, its still trained on the data it's fed