Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
[dead]
6 months ago
|
hide
|
past
|
favorite
luis_likes_math
6 months ago
[–]
Hello! Here is a breakdown of GRPO (Group Relative Policy Optimization), used to train reasoning models like DeepSeek.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
