Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
gldjmp
on Jan 5, 2025
|
parent
|
context
|
favorite
| on:
30% drop in O1-preview accuracy when Putnam proble...
You can get all of the correct numerical answers for the Putnam and still get a zero, because the reasoning is graded very harshly. The scores measured in this paper are not comparable to actual Putnam scores.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: