IDEAS has been a great partner at pushing AI for sciences and it has developed into an exceptional powerhouse of European innovation.
Now, I find it very quite troubling that there was no transparency about the recent leadership selection process and I have mostly heared
Deeply concerned about the future of IDEAS NCBR, a pioneering AI research institute & newly approved ELLIS unit. International advisory board members, incl. Aleksander MΔ dry, have resigned after the sudden and unjustified replacement of its founding director. @donaldtuskEPP
Tried @grok 4 on a dozen non-trivial math (under/)grad level math problems.
So far, it has failed to fail me even once.
Congrats to @Yuhu_ai_, @ericzelikman and the whole xAI reasoning team, their progress has exceeded all my expectation!
A few months ago, I had a long chat with Terry Tao. While he is one of the knowledgeabe and forward-looking mathematician, I still had the impression that he was out of touch withe the reality that was coming up quickly.
Deep learning is not just memorization, it also compresses. And really good compression amounts to intelligence. That's why we get few/zero-shot capabilities from LLMs, since they "discovered" a lot of the underlying patterns just by being trained to compress.
Deep learning takes data points and turns them into a query-able structure that enables retrieval and interpolation between the points.
You could think of it as a continuous generalization of database technology.
New jobs in the 21st century:
Model restart specialist
Hyperparameter psychic
Prompt engineer
Model janitor
Tensor shape mediator
Quantum state observer
Model footprint accountant
I talked to several non-ML people over the past half year (including my kids) about chat bots and what i feel is a general frustration about their being moralizing and not getting answers even to relatively innocent prompts. So being more unhinged will be a feature not a bug.
User: "When was Einstein Born?"
LLM: .. Let's revisit a compressed form of all existing knowledge on the web ... once for *every* single input token ... and each token generated, including the GDP of Armenia, the name of all kings ever lived and all famous chess games, etc ...
One on math: super-human mathematician AI by June 26. The other: 95% at the end of 2025 on the current test set of of ARC.
We can work out the details on the first. I feel much less confident about the second, esp. given the current challanges with evaluation on proprietary
Here is a thread in paper references demonstrating that current deep learning, especially transformers are amazingly powerful for symbol manipulation already:
@GuillaumeLample, @f_chartonarxiv.org/abs/1912.01412
Solving hard integrals using deep transformers
π§΅1/n
Neural networks have a reputation for being better at solving statistical or approximate problems than at performing calculations or working with symbolic data. In this paper, we show that they...
AI Agenda: Why Muskβs AI Rivals Are Alarmed by His New GPU Cluster
Elon Musk claims to have finished a 100,000-strong H100 cluster in four months. How likely is that?
theinformation.com/articles/why-mβ¦
From @anissagardizy8
My view on this has not changed in the past eight years: I have given many talks and written position paper in 2019 (link below). Progress is faster than my past expectation. My target date used to be ~2029 back then. Now it is 2026 for a superhuman AI mathematician. While a