Terence TaoSep 6, 2025

Earlier this year, the second #AIMO (artificial intelligence mathematical olympiad) concluded, with the winning team solving 34/50 in the final set of math problems (that had been selected to be harder for AI than the first AIMO).

The competition was restricted to open source models and run with a limited amoutn of compute. The AIMO has now conducted a retest of these problems both for the top two teams from that competition (NemoSkills and imagination research), as well as OpenAI's o3 model, both with comparable levels of compute resources, and with high resources. Unsurprisingly, the high resource models did better, with the high resource o3 model scoring as high as 47/50, or even 50/50 if given two tries at each question. On the other hand, the gap between the open source models and the commercial models for a fixed amount of compute was relatively slight.

More details of this experiment are available at https://aimoprize.com/updates/2025-09-05-the-gap-is-shrinking

The gap between commercial and open-source LLMs for Olympiad-level math is shrinking | AIMO Prize

AIMO Prize
Afro House KingApr 27, 2025

Aimo – Hopeful Thoughts
https://afrohouseking.com/aimo-hopeful-thoughts/

#AfroHouse2025 #Aimo

Afro House KingApr 6, 2025

Aimo, Sabuddha Kortez, Naux – Let It Rain (2025 Reblast)
https://afrohouseking.com/aimo-sabuddha-kortez-naux-let-it-rain-2025-reblast/

#AfroHouse2025 #Aimo

Aimo, Sabuddha Kortez, Naux - Let It Rain (2025 Reblast) • DOWNLOAD MP3

Aimo, Sabuddha Kortez, Naux - Let It Rain (2025 Reblast), PRIMARY DOWNLOAD, • Afro House King

Afro House King
José A. AlonsoNov 30, 2024
Readings shared November 29, 2024. https://jaalonso.github.io/vestigium/posts/2024/11/29-readings_shared_11-29-24 #ITP #Lean4 #IsabelleHOL #Logic #Math #Calculemus #MachineLearning #LLMs #AIMO #AlphaProof #AlphaGeometry #Haskell #FunctionalProgramming #Programming #FormalVerification #Dafny
Readings shared November 29, 2024

The readings shared in Bluesky on 29 November 2024 are Readings shared November 28, 2024. #ITP #IsabelleHOL #Logic #Math #Haskell #FunctionalProgramming Proof of "If the limit of the sequence u(n) is

Vestigium
Terence TaoNov 29, 2024
The second Artificial Intelligence Mathematical Olympiad #AIMO challenge, which has been ongoing for about a month, is closing in on the 20/50 threshold (on the public leaderboard) needed to trigger the "Early Sharing Prize" for the first contestant to reach this threshold while sharing their notebook. Currently, the best submissions are at 18/20: https://www.kaggle.com/competitions/ai-mathematical-olympiad-progress-prize-2/leaderboard . A key input to recent progress was a new whitelisting process https://www.kaggle.com/competitions/ai-mathematical-olympiad-progress-prize-2/discussion/548129 for models (led by @friederrr), leading to the approval of a powerful new open source model - a specific instance of QwQ-32B - which seems to be significantly better at these types of competition problems than previous open source models.
AI Mathematical Olympiad - Progress Prize 2

Solve national-level math challenges using artificial intelligence models

José A. AlonsoNov 29, 2024
Machine learning and mathematics (The twin catalysts poised to accelerate twenty-first century mathematics: formalisation and machine learning). ~ Harald Carlens. https://joltml.com/ml-mathematics/ #Math #ITP #Lean4 #MachineLearning #LLMs #AIMO #AlphaProof #AlphaGeometry
Machine Learning and Mathematics

The twin catalysts poised to accelerate twenty-first century mathematics: formalisation and machine learning.

Terence TaoOct 17, 2024
The second #AIMO progress prize has now been launched: https://www.kaggle.com/competitions/ai-mathematical-olympiad-progress-prize-2 . The format is similar to the previous progress prize at https://www.kaggle.com/c/ai-mathematical-olympiad-prize , but the problems are harder (to reflect the improved baseline state of the art arising from the previous prize), slightly more general (in the first prize, answers were a three digit number, now they are instead labeled with a three digit number), hardware for evaluation has been upgraded, and the monetary prizes have been doubled. Submissions close on Apr 1, 2025.
AI Mathematical Olympiad - Progress Prize 2

Solve national-level math challenges using artificial intelligence models

Terence TaoJul 4, 2024
Preliminary results for the first #AIMO progress prize is up: https://www.kaggle.com/competitions/ai-mathematical-olympiad-prize/leaderboard Provisionally at least, it appears that the winning program managed to correctly answer 29/50 of the private test questions, which is somewhat higher than expected.
AI Mathematical Olympiad - Progress Prize 1

Solve national-level math challenges using artificial intelligence models

José A. AlonsoApr 19, 2024
Symbolic computation for all the fun. ~ Chad E. Brown, Mikoláš Janota, Mirek Olšák. https://arxiv.org/abs/2404.12048 #AIMO #ATP #SMT #Math
Symbolic Computation for All the Fun

Motivated by the recent 10 million dollar AIMO challenge, this paper targets the problem of finding all functions conforming to a given specification. This is a popular problem at mathematical competitions and it brings about a number of challenges, primarily, synthesizing the possible solutions and proving that no other solutions exist. Often, there are infinitely many solutions and then the set of solutions has to be captured symbolically. We propose an approach to solving this problem and evaluate it on a set of problems that appeared in mathematical competitions and olympics.

arXiv.org
Simon StrandgaardApr 3, 2024

AIMO. Crazy development already after 2 days since the contest started.

Now there is a new 1st place that solves 5 of the tasks!
https://www.kaggle.com/competitions/ai-mathematical-olympiad-prize/leaderboard
#aimo #ai #math #olympics #agi #kaggle

AI Mathematical Olympiad - Progress Prize 1

Solve national-level math challenges using artificial intelligence models