BigMuffin69 @ BigMuffin69 @awful.systems Posts 6Comments 89Joined 1 yr. ago
Remember how OAI claimed that O3 had displayed superhuman levels on the mega hard Frontier Math exam written by Fields Medalist? Funny/totally not fishy story haha. Turns out OAI had exclusive access to that test for months and funded its creation and refused to let the creators of test publicly acknowledge this until after OAI did their big stupid magic trick.
From Subbarao Kambhampati via linkedIn:
"๐๐ง ๐ญ๐ก๐ ๐ฌ๐๐๐๐ฒ ๐จ๐ฉ๐ญ๐ข๐๐ฌ ๐จ๐ "๐ฉ๐๐๐๐ ๐๐๐ ๐๐ ๐จ๐ฎ๐ฐ ๐ด๐๐๐ ๐๐ ๐ช๐๐๐๐๐๐๐๐๐ ๐ฉ๐๐๐๐๐๐๐๐ ๐ช๐๐๐๐๐๐๐" hashtag#SundayHarangue. One of the big reasons for the increased volume of "๐๐๐ ๐๐จ๐ฆ๐จ๐ซ๐ซ๐จ๐ฐ" hype has been o3's performance on the "frontier math" benchmark--something that other models basically had no handle on.
We are now being told (https://lnkd.in/gUaGKuAE) that this benchmark data may have been exclusively available (https://lnkd.in/g5E3tcse) to OpenAI since before o1--and that the benchmark creators were not allowed to disclose this *until after o3 *.
That o3 does well on frontier math held-out set is impressive, no doubt, but the mental picture of "๐1/๐3 ๐๐๐๐ ๐๐๐๐ ๐๐๐๐๐ ๐๐๐๐๐๐๐ ๐๐ ๐๐๐๐๐๐ ๐๐๐๐, ๐๐๐ ๐๐๐๐ ๐๐๐๐๐๐๐๐๐๐๐๐ ๐๐๐๐๐๐๐๐๐๐ ๐๐ ๐๐๐๐๐๐๐๐ ๐๐๐๐"--that the AGI tomorrow crowd seem to have--that ๐๐ฑ๐ฆ๐ฏ๐๐ ๐ธ๐ฉ๐ช๐ญ๐ฆ ๐ฏ๐ฐ๐ต ๐ฆ๐น๐ฑ๐ญ๐ช๐ค๐ช๐ต๐ญ๐บ ๐ค๐ญ๐ข๐ช๐ฎ๐ช๐ฏ๐จ, ๐ค๐ฆ๐ณ๐ต๐ข๐ช๐ฏ๐ญ๐บ ๐ฅ๐ช๐ฅ๐ฏ'๐ต ๐ฅ๐ช๐ณ๐ฆ๐ค๐ต๐ญ๐บ ๐ค๐ฐ๐ฏ๐ต๐ณ๐ข๐ฅ๐ช๐ค๐ต--is shattered by this. (I have, in fact, been grumbling to my students since o3 announcement that I don't completely believe that OpenAI didn't have access to the Olympiad/Frontier Math data before hand.. )
I do think o1/o3 are impressive technical achievements (see https://lnkd.in/gvVqmTG9 )
๐ซ๐๐๐๐ ๐๐๐๐ ๐๐ ๐๐๐๐ ๐๐๐๐๐๐๐๐๐๐ ๐๐๐๐ ๐๐๐ ๐๐๐ ๐๐๐๐๐ ๐๐๐๐๐๐ ๐๐ ๐๐ ๐๐๐๐๐ ๐๐๐๐๐๐๐๐๐๐--๐๐๐ ๐ ๐๐๐๐'๐ ๐๐๐๐๐ ๐๐๐๐๐๐ "๐จ๐ฎ๐ฐ ๐ป๐๐๐๐๐๐๐."
We all know that data contamination is an issue with LLMs and LRMs. We also know that reasoning claims need more careful vetting than "๐ธ๐ฆ ๐ฅ๐ช๐ฅ๐ฏ'๐ต ๐ด๐ฆ๐ฆ ๐ต๐ฉ๐ข๐ต ๐ด๐ฑ๐ฆ๐ค๐ช๐ง๐ช๐ค ๐ฑ๐ณ๐ฐ๐ฃ๐ญ๐ฆ๐ฎ ๐ช๐ฏ๐ด๐ต๐ข๐ฏ๐ค๐ฆ ๐ฅ๐ถ๐ณ๐ช๐ฏ๐จ ๐ต๐ณ๐ข๐ช๐ฏ๐ช๐ฏ๐จ" (see "In vs. Out of Distribution analyses are not that useful for understanding LLM reasoning capabilities" https://lnkd.in/gZ2wBM_F ).
At the very least, this episode further argues for increased vigilance/skepticism on the part of AI research community in how they parse the benchmark claims put out commercial entities."
Big stupid snake oil strikes again.
Lmaou. "We need to alignment pill the Russian youth." Fast forward to the year 20XX and the haunted alignment pilled adults are now 'aligning' their bots to the world's top nuclear armed despot.
tony_soprano_how_could_this_happen.jpg (for some reason awful systems won't let me upload pictures anymore (ใเฒ ็เฒ )ใ)
Holy Moses in heaven, iirc both Sam and Dario have said that their urge to build the torment nexus came from being inspired by online RAT forums. Maybe alignment 'pilling' youths is counterproductive to human flourishing? As the LWers say, "update your priors fuckheads"
smh they really do be out here believing there's a little man in the machine with goals and desires, common L for these folks
ong Yann LeCun was sharing this post too and i was shook that he was seeing quality shit post like this before me. We are not ready for whats coming next . jpg
Fellas, I was promised the first catastrophic AI event in 2024 by the chief doomers. There's only a few hours left to go, I'm thinking skynet is hiding inside the times square orb. Stay vigilant!
The ARC scores don't matter too much to me at 3k a problem. Like the original goal of the prize had a compute limit. You can't break that rule and then claim victory ( I mean I guess you can, but like not everyone is gonna be as wowed as xitter randos, ensemble methods were already hitting 80% + acc to francois )
And unfortunately, with Frontier math, the lack of transparency w.r.t. which problems were solved and how they were solved makes it frustrating as hell to me, as someone who actually would like to see a super math robot. According to the senior math advisor to the people who created the data set, iirc 40% solved problems were in the easiest category / 50% in the second tier category and 10% were in the "hard" tier, but he said that he looked at the solutions and that they looked like mostly being solved 'heuristically' instead of plopping out any 'new' insights.
Again, none of this is good science, just pure shock and awe. I've heard rumors that OAI is hiring strong competition style mathematicians to supervise the reinforcement learning for these types of problems and if they are letting O3 take the test, then how the hell does that not leak the problem set? Like now the whole test is compromised now right? Since this behemoth uses enough electricity to power a city block, theres no way they would be able to run it locally. Now OAI can literally pay their peeps to solve the rest and surprise surprise O3++ will hit 80%
OTOH, with code forces scores and math scores this high, I can now put on my LW cap and say this model has 2 trillion IQ, so why hasn't it exterminated me and my family yet like big Yud promised? It's almost as if there is no little creature inside trying to take over the world or something.
but muh "nice sneers for winners" ;_;
Thank you. My wife is deathly allergic to shrimp, and I live by the motto
'If they send one of your loved ones to the emergency room, you send 10 of theirs to the deep fryer. '
Shared this on tamer social media site and a friend commented:
"That's nonsense. The largest charities in the country are Feeding America, Good 360, St. Jude's Children's Research Hospital, United Way, Direct Relief, Salvation Army, Habitat for Humanity etc. etc. Now these may not satisfy the EA criteria of absolutely maximizing bang for the buck, but they are certainly mostly doing worthwhile things, as anyone counts that. Just the top 12 on this list amount to more than the total arts giving. The top arts organization on this list is #58, the Metropolitan Museum, with an income of $347M."
Wild. Just the mention of "the moon" and it starts playing in my head. This place is an info hazard.
Pat walking into the last board meeting
Same reaction. I don't see how you can stop at it's immoral for me to not breed with the รbermensch when the clear logical end is it's immoral for me to breed entirely. As a childless person, do I get to go to EA meet ups and wag my fingers at them for not being moral now?
I will find someone who I consider better than me in relevant ways, and have them provide the genetic material. I think that it would be immoral not to, and that it is impossible not to think this way after thinking seriously about it.
Corporate needs you to find the difference between this^ and our local cult leader is the sun god reborn, it's every woman's duty to carry his seed. It is immoral to deny his divine will.
I remember when several months (a year ago?) when the news got out that gpt-3.5-turbo-papillion-grumpalumpgus could play chess around ~1600 elo. I was skeptical the apparent skill wasn't just a hacked-on patch to stop folks from clowning on their models on xitter. Like if an LLM had just read the instructions of chess and started playing like a competent player, that would be genuinely impressive. But if what happened is they generated 10^12 synthetic games of chess played by stonk fish and used that to train the model- that ain't an emergent ability, that's just brute forcing chess. The fact that larger, open-source models that perform better on other benchmarks, still flail at chess is just a glaring red flag that something funky was going on w/ gpt-3.5-turbo-instruct to drive home the "eMeRgEnCe" narrative. I'd bet decent odds if you played with modified rules, (knights move a one space longer L shape, you cannot move a pawn 2 moves after it last moved, etc), gpt-3.5 would fuckin suck.
Edit: the author asks "why skill go down tho" on later models. Like isn't it obvious? At that moment of time, chess skills weren't a priority so the trillions of synthetic games weren't included in the training? Like this isn't that big of a mystery...? It's not like other NN haven't been trained to play chess...
If they do press conferences this time around, ever question should just be "does Elon approve of decision ____ ?" Will drive Trump fkn insane.