Start Your NewsReadery Pro FREE TRIAL!
Register and verify your email address to start your
NewsReadery Pro FREE TRIAL
today!
Login / Register
Actions
Newsfeed
Popular
Local
Discover
Follows
Blocks
Bookmarks
Filters
Settings
PowerUps
Help
Updates
Login / Register
arstechnica.com
/
Share Newsitem
View, share or embed this newsitem using the details below.
Ars Technica
@arstechnica.com
/
40 days
40d
Follow
New secret math benchmark stumps AI models and PhDs alike
FrontierMath's difficult questions remain unpublished so that AI companies can't train against it.
https://newsreadery.com/go/9c1b2f1d48b301624e9cab27ca77a38c
Save
Share
Share this Newsitem URL
Embed this Newsitem
<div class="readery-embed readery-embed-9c1b2f1d48b301624e9cab27ca77a38c"></div><script async src="https://open.newsreadery.com/oembed?format=html&url=https%3A%2F%2Fnewsreadery.com%2Fitem%2F9c1b2f1d48b301624e9cab27ca77a38c" charset="utf-8"></script>
Back
Read Full Article
Continue
Please wait ...
