Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
PoignardAzur
on March 25, 2023
|
parent
|
context
|
favorite
| on:
GPT-4 performs significantly worse on coding probl...
The most naive way you could do things could be to procedurally generate immense amounts of python code, then ask the model to predict whether the code will compile, whether it will crash, what its outputs will be given certain inputs, etc.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: