Dev

Using GRPO to Beat o1, o3-mini and R1 at “Temporal Clue”

Article URL: https://openpipe.ai/blog/using-grpo-to-beat-o1-o3-mini-and-r1-on-temporal-clue Comments URL: https://news.ycombinator.com/item?id=43284420 Points: 179 # Comments: 46

Mar 7, 2025 - 17:42
 0
Using GRPO to Beat o1, o3-mini and R1 at “Temporal Clue”

Article URL: https://openpipe.ai/blog/using-grpo-to-beat-o1-o3-mini-and-r1-on-temporal-clue

Comments URL: https://news.ycombinator.com/item?id=43284420

Points: 179

# Comments: 46

Lire plus

Tags :

Article précédent

Matters Computational (2010) [pdf]

Article suivant

Quel est l’emploi du temps le plus efficace pour une pompe de piscine ?

Articles similaires

In Defense of Text Labels: Why Icons Alone Aren't Enough

In Defense of Text Labels: Why Icons Alone Aren't Enough

Fév 23, 2025  0

Ploomber (YC W22) Is Hiring Engineers (Infra, Backend, Growth) and Ex-Founders

Ploomber (YC W22) Is Hiring Engineers (Infra, Backend, ...

Fév 5, 2025  0

Kaneo – An open source project management platform

Kaneo – An open source project management platform

Fév 22, 2025  0