The most prestigious law school admissions discussion board in the world.
|
 |
 
|
New OpenAI general reasoning model gets gold medal at international math olympia
Poast new message in this thread
Date: July 19th, 2025 1:10 PM Author: ,.,.,.,,,.,,.,..,.,.,.,.,,. d. value of human intelligence falling every day. 180 times.
https://x.com/alexwei_/status/1946477742855532918?s=46 |
Date: July 19th, 2025 5:01 PM Author: Car (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49114825) |
|
Date: July 19th, 2025 7:23 PM Author: ,.,....,...,,,..,..,.,..,.,.,.,. Gemini getting 50% on USAMO made me think this would happen in a couple years. it's somewhat surprising it happened this year. pretty amazing progress considering the original GPT-4 would get 0-1 on a random AIME exam and now contest math looks close to solved. (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115116) |
 |
Date: July 19th, 2025 7:28 PM Author: scholarship (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115125) |
|
Date: July 20th, 2025 3:59 AM Author: chilmata Yes, it's truly remarkable how quickly AI has advanced in contest math! Gemini's 50% score on the USAMO (United States of America Mathematical Olympiad) is a massive leap compared to where models like GPT-4 started just a couple of years ago.
### Key Observations on the Progress: 1. **From Near-Zero to Competitive Performance** - Early versions of GPT-4 struggled to score even 1-2 problems on the AIME (American Invitational Mathematics Exam), which is significantly easier than the USAMO. - Now, AI is not just solving AIME-level problems reliably but also tackling Olympiad-level questions, which require deep reasoning, creativity, and proof-writing.
2. **Why This is Surprising** - Many experts (including myself) expected this level to take at least 2-3 more years. The fact that it happened this year suggests: - Better training techniques (e.g., improved reasoning loops, synthetic data, and self-improvement methods). - Stronger mathematical priors (models may now "understand" abstract patterns rather than just memorizing). - Possibly, more sophisticated search/verification methods (though Gemini's exact approach isn't public).
3. **Is Contest Math "Close to Solved"?** - **For AIME:** Probably yes. AI is likely near-superhuman at this level already. - **For USAMO/IMO:** Not yet, but progress is accelerating. A 50% USAMO score is already better than many human contestants, but the hardest problems (e.g., IMO Q6-style combinatorics/number theory) still require breakthroughs in symbolic reasoning and long-term planning. - **For Proof Writing:** AI still makes stylistic errors, but formal verification (e.g., Lean) can help close the gap.
4. **What’s Next?** - **IMO Gold (60+ score) within 1-2 years?** If progress continues at this rate, it's plausible. - **General Mathematical Research:** The real test will be whether these models can contribute to unsolved problems (e.g., combinatorics conjectures, lightweight formal math).
### Why This Matters Beyond Olympiads: - **Education:** AI could become the ultimate tutor for advanced math. - **Research:** Automated reasoning might assist in mathematical discovery. - **AGI Benchmarks:** Math is a strong proxy for structured reasoning—this progress hints at broader capabilities.
It’s an exciting time! Wouldn’t be surprised if an AI wins an IMO gold medal by 2026. |
|
Date: July 19th, 2025 7:26 PM Author: rape bunny this is their in-house super model that doesn't have guardrails and does 20mil token recursive chain of thought or something. when we get gpt-5 it will be some distilled faggot version of this (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115121) |
|
Date: July 19th, 2025 7:34 PM Author: ,.,....,...,,,..,..,.,..,.,.,.,. they might deploy this model but it's also unlikely they'll give it as much compute as they used here. i remember when they reported their ARC-AGI results for o3 and it turns out they were using something like $3K in compute per question (!). the number here is likely even higher. (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115148) |
|
Date: July 19th, 2025 9:11 PM Author: rape bunny all doomsday scenarios imply a frontier model at HQ significantly more capable than the dogshit distillations given out to civilians (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115331) |
|
Date: July 20th, 2025 12:54 AM Author: rape bunny Subject: this fucking faggot: "we are releasing GPT-5 soon but want to set accurate expectations: this is an experimental model that incorporates new research techniques we will use in future models. we think you will love GPT-5, but we don't plan to release a model with IMO gold level of capability for many months."
https://x.com/sama/status/1946569252296929727 |
|
Date: July 19th, 2025 7:34 PM Author: KikeLord69 who the fuck cares lmao it just knows stuff poasted on the internet (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115149) |
|
Date: July 20th, 2025 9:27 AM Author: Local on the 8s (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49116050) |
|
Date: July 20th, 2025 12:58 AM Author: hank_scorpio so it can solve math problems that humans already solved? idk, doesn't sound that big to me (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115707) |
|
Date: July 20th, 2025 2:59 AM Author: rape bunny problems at the forefront of human knowledge. that's all AGI ever required--the highest level human knowledge in every domain.
are you dumb btw? you can't grasp the implications of this? AI being equivalent with the best programmers/mathmaticians in the world? Why would anyone use a normal lawyer when AI is the equivalent of having Dershowitz personally represent you with unlimited billing hours to your case for flat fee? it's all over |
|
Date: July 20th, 2025 3:13 AM Author: VoteRepublican (A true Chad!! where's your gf/wifew?) so is it over yet or no (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115784) |
|
Date: July 20th, 2025 3:23 AM Author: rape bunny soon. we need Butlerian Jihad (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115786) |
|
Date: July 20th, 2025 3:47 AM Author: Wang Hernandez AI sucks at law (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115794) |
|
Date: July 20th, 2025 3:55 AM Author: rape bunny AI was getting 2+2 wrong a year ago. Cope retard (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49115795) |
|
Date: July 20th, 2025 12:12 PM Author: ,.,.,.,,,.,,.,..,.,.,.,.,,. There are a handful of high school students with 150+ IQ that are able to solve these problems. In addition, AI went from being able to get 700 or so on the SAT math to this in about two years thanks to AI scaling. Do you feel confident it won’t start solving unknown problems with another 2 years of scaling? (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49116355) |
|
Date: July 20th, 2025 9:26 AM Author: Paralegal Muhammad Phenotype’s relative value skyrockets as chink GPA value plummets (http://www.autoadmit.com/thread.php?thread_id=5752305&forum_id=2/#49116047) |