Basically a boid simulation where a swarm of birds can collectively solve MNIST. The goal is not some new SOTA architecture, it is to find the right trade-off where the system already exhibits complex emergent behavior while the swarming rules are still simple.
It is currently abandoned due to a serious lack of free time (*), but I would consider collaborating with anyone willing to put in some effort.
100% frontpage-worthy! Frankly I was already bored with all those pelicans, and a bit worried that the labs are overfitting on pelicans specifically. This test clearly demonstrates that they are not.
That's very cool, but it's not an apples to apples comparison. The reasoning model learned how to do long multiplication. (Either from the internet, or from generated examples of long multiplication that were used to sharpen its reasoning skills. In principle, it might have invented it on its own during RL, but no, I don't think so.)
In this paper, the task is to learn how to multiply, strictly from AxB=C examples, with 4-digit numbers. Their vanilla transformer can't learn it, but the one with (their variant of) chain-of-thought can. These are transformers that have never encountered written text, and are too small to understand any of it anyway.
If being probabilistic prevented learning deterministic functions, transformers couldn’t learn addition either. But they can, so that can't be the reason.
Are you sure? I bet you if you pull 10 people off the street and ask them to multiply 5 digit by 5 digit numbers by hand, you won't have a 100% success rate.
My uninformed and perhaps overly charitable interpretation: he warned them they were going to be steamrolled, they built their product anyway, and now OpenAI is buying them because (1) OpenAI doesn't want the negative publicity of steamrolling them all, and (2) OpenAI has the money and is a bit too lazy to build a clone.
There's a related section about 'mathiness' in section 3.3 of the article "Troubling Trends in Machine Learning Scholarship" https://arxiv.org/abs/1807.03341. I would say the situation has only gotten worse since that paper was written (2018).
However the discussion there is more about math which is unnecessary to a paper, not so much about the problem of math which is unintelligible or, if intelligible, then incorrect. I don't have other papers off the top of my head, although by now it's my default expectation when I see a math-centric AI paper. If you have any such papers in mind, I could tell you my thoughts on it.
You dismiss parent's example test because it's in the training data. I assume you also dismiss the Sally-Ann test, for the same reason. Could you please suggest a brand new test not in the training data?
FWIW, I tried to confuse 4o using the now-standard trick of changing the test to make it pattern-match and overthink it. It wasn't confused at all:
I can't suggest a new test no, it is a hard problem and identifying problems is usually easier than solving them.
I'm just trying to say that strong claims require strong evidence, and a claim that LLM's can have theory of mind and thus "understand that other people have different beliefs, desires, and intentions than you do" is a very strong claim.
It's like giving students the math problem of 1+1=2 and loads of examples of it solved in front of them, and then testing them on you have 1 apple, and I give you another apple, how many do you have, and then when they are correct saying that they can do all additive based arithmetic.
This is why most benchmark tests have many many classes of examples, for example looking at current theory of mind benchmarks [1], we can see slightly more up to date models such as o1-preview still scoring substantially below human performance. More importantly by simply changing the perspective from first to third person, accuracy drops in LLM models by 5-15% (percent score, not relative to its performance), whilst it doesn't change for human participants, which tells you that something different is going on there.
https://github.com/danielvarga/transformer-as-swarm
Basically a boid simulation where a swarm of birds can collectively solve MNIST. The goal is not some new SOTA architecture, it is to find the right trade-off where the system already exhibits complex emergent behavior while the swarming rules are still simple.
It is currently abandoned due to a serious lack of free time (*), but I would consider collaborating with anyone willing to put in some effort.
(*) In my defense, I’m not slacking meanwhile: https://arxiv.org/abs/2510.26543 https://arxiv.org/abs/2510.16522 https://www.youtube.com/watch?v=U5p3VEOWza8
reply