Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

comparing their benchmark scores is fucking nuts bro

LLMenjoyer's profile picture

llm_enjoyer

1,379 subscribers

33,103 Aufrufe • vor 7 Monaten •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Lacey is nuts bro😭
0:17

Lacey is nuts bro😭

E

504,036 Aufrufe • vor 9 Monaten

Rayan Cherki’s vision is actually nuts bro
0:11

Rayan Cherki’s vision is actually nuts bro

Danis

824,705 Aufrufe • vor 1 Monat

Laliga Camera quality is NUTS HOLY FUCKING AURAAAA!!!!!
0:11

Laliga Camera quality is NUTS HOLY FUCKING AURAAAA!!!!!

َ

1,574,296 Aufrufe • vor 23 Tagen

BRO, HECHICERO IS FUCKING INCREDIBLE. #ViernesEspectacularCMLL
0:11

BRO, HECHICERO IS FUCKING INCREDIBLE. #ViernesEspectacularCMLL

Drainmaker

18,517 Aufrufe • vor 8 Monaten

Bro Chet is 7 fucking feet
0:16

Bro Chet is 7 fucking feet

Mike

1,284,547 Aufrufe • vor 7 Monaten

THE NUTS ON DYLAN FUCKING HARPER
0:11

THE NUTS ON DYLAN FUCKING HARPER

Bala

1,299,489 Aufrufe • vor 4 Monaten

bro this hoseok is a fucking PROBLEM.
0:14

bro this hoseok is a fucking PROBLEM.

minie

86,921 Aufrufe • vor 2 Monaten

Lil bro is a fucking THUG 😭
0:15

Lil bro is a fucking THUG 😭

internet hall of fame

8,959,781 Aufrufe • vor 1 Jahr

Academic faculty comparing their research output 😅
0:17

Academic faculty comparing their research output 😅

Kosta Derpanis

302,898 Aufrufe • vor 3 Jahren

this year is so fucking nuts bc tell me how we're getting kate mckinnon as a gay scientist again and her partner is geena fucking davis
0:12

this year is so fucking nuts bc tell me how we're getting kate mckinnon as a gay scientist again and her partner is geena fucking davis

margot 🐀

79,073 Aufrufe • vor 3 Monaten

bro the sound design is so fucking cool
0:10

bro the sound design is so fucking cool

d0nut2x 🐀

79,851 Aufrufe • vor 8 Monaten

BRO THIS ANIMATED SEQUENCE IS SO FUCKING COOL
0:14

BRO THIS ANIMATED SEQUENCE IS SO FUCKING COOL

yo boi K

174,592 Aufrufe • vor 2 Monaten

Flow state rashford fucking nuts man😭
0:15

Flow state rashford fucking nuts man😭

AB⚕

168,216 Aufrufe • vor 6 Monaten

💻Tired of running so many slow, expensive benchmark evals across every checkpoint? Try ✨BenchPress✨ at provide a few benchmark scores, then get predictions for the remaining ~100 benchmarks, with trust probabilities and calibrated 90% prediction intervals. How does this work? In his original post ( Dimitris Papailiopoulos first tried the idea as a fun question: collect model-by-benchmark scores into a matrix, find its low-rank structure, and use matrix completion to predict missing benchmark scores from a few observed ones. We expanded this into a full system: a fully audited 84-model x 133-benchmark score matrix, an optimized matrix-completion predictor, and a reliability layer for trust probabilities and 90% prediction intervals. Beyond predicting missing scores, we also suggest practical seed benchmark sets. The five-probe set {GPQA-D, HLE, Codeforces, MMLU-Pro, ARC-AGI-1} recovers the rest of a model's public score profile with a MedAE of 3.93 points. A lower-cost set {GPQA-D, MMLU-Pro, Aider Polyglot, MATH-500, AIME 2026} reaches 4.55 points. See more details below 🧵1/7 This work is with Dimitris Papailiopoulos at AI Frontiers, a boutique research lab inside Microsoft Research.
0:14

💻Tired of running so many slow, expensive benchmark evals across every checkpoint? Try ✨BenchPress✨ at provide a few benchmark scores, then get predictions for the remaining ~100 benchmarks, with trust probabilities and calibrated 90% prediction intervals. How does this work? In his original post ( Dimitris Papailiopoulos first tried the idea as a fun question: collect model-by-benchmark scores into a matrix, find its low-rank structure, and use matrix completion to predict missing benchmark scores from a few observed ones. We expanded this into a full system: a fully audited 84-model x 133-benchmark score matrix, an optimized matrix-completion predictor, and a reliability layer for trust probabilities and 90% prediction intervals. Beyond predicting missing scores, we also suggest practical seed benchmark sets. The five-probe set {GPQA-D, HLE, Codeforces, MMLU-Pro, ARC-AGI-1} recovers the rest of a model's public score profile with a MedAE of 3.93 points. A lower-cost set {GPQA-D, MMLU-Pro, Aider Polyglot, MATH-500, AIME 2026} reaches 4.55 points. See more details below 🧵1/7 This work is with Dimitris Papailiopoulos at AI Frontiers, a boutique research lab inside Microsoft Research.

Yuchen Zeng

36,680 Aufrufe • vor 3 Monaten

Comparing Messi with Ronaldo in 2026 come on bro it’s no longer a debate
0:14

Comparing Messi with Ronaldo in 2026 come on bro it’s no longer a debate

Storms

27,756 Aufrufe • vor 3 Monaten