niftycent.com
NiftyCent
RO
Logare
Piata de desfacere
Reduceri

30% Drop In o1-Preview Accuracy When Putnam Problems Are Slightly Variated

Article URL: https://openreview.net/forum?id=YXnwlZe0yf&noteId=yrsGpHd0Sf

Comments URL: https://news.ycombinator.com/item?id=42565606

Points: 61

# Comments: 30

https://openreview.net/forum?id=YXnwlZe0yf&noteId=yrsGpHd0Sf

Creată 3d | 1 ian. 2025, 13:40:06


Autentifică-te pentru a adăuga comentarii

Alte posturi din acest grup

RailsConf 2025 will be the last gathering of its kind
RailsConf 2025 will be the last gathering of its kind

Article URL: https://rubycentral.org/news/announcing-railsconf-2025-and-a-new-chapter-fo

4 ian. 2025, 06:30:18 | Hacker news
Do Files want to be Actors?
Do Files want to be Actors?

Article URL: https://lewiscampbell.tech/blog/250104.html

Comments URL: http

4 ian. 2025, 06:30:16 | Hacker news
Breaking Up with Long Tasks or: how I learned to group loops and wield the yield
Breaking Up with Long Tasks or: how I learned to group loops and wield the yield

Article URL: https://calendar.perfplanet.com/2024/breaking-up-with-l

4 ian. 2025, 06:30:14 | Hacker news
XORShift for Magic Bitboards
XORShift for Magic Bitboards

Article URL: https://www.strydr.net/articles/devlog-0x1

Comments URL: https:

4 ian. 2025, 06:30:14 | Hacker news
U-Haul Growth Metros and Cities of 2024: Dallas Top Metro for In-Migration
U-Haul Growth Metros and Cities of 2024: Dallas Top Metro for In-Migration

Article URL: https://www.uhaul.com/Articles/About/U-Haul-Growth-Metr

4 ian. 2025, 06:30:11 | Hacker news
Wine 10.0 RC4 – Run Windows Applications on Linux, BSD, Solaris and macOS
Wine 10.0 RC4 – Run Windows Applications on Linux, BSD, Solaris and macOS

Article URL: https://gitlab.winehq.org/wine/wine/-/releases/wine-10.0-rc4

Comments URL:

4 ian. 2025, 06:30:10 | Hacker news
China is the sole manufacturing superpower
China is the sole manufacturing superpower

Article URL: https://cepr.org/voxeu/columns/china-worlds-sole-manufacturing-superpower-line-ske

4 ian. 2025, 06:30:06 | Hacker news
Techie
Techie