Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
wouldbecouldbe
18 days ago
|
parent
|
context
|
favorite
| on:
OpenAI claims gold-medal performance at IMO 2025
Yeah that’s the dream, but same as with the bar exams, they are fine tuning the models for specific tests. Which probably the model even has been trained on previous version of those tests
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: