logoalt Hacker News

andriy_kovalyesterday at 7:49 PM2 repliesview on HN

I think it could indicate that "semi-private" dataset likely leaked to their training data.


Replies

xpctyesterday at 8:47 PM

A dataset being as popular as their's is will contaminate the data just by people discussing it and creating their own public test sets of similar problems.

Still, probably not that much compared to employees targeting it.

IshKebabyesterday at 7:56 PM

It says "Provider Adapter" so presumably they put some manual work in to make this work.