Reposted by Samuel MüllerJeff Dean @jeffdean.bsky.social · 09/01/2026The recent days have been horrific. We can't become numb to repeated instances of illegal and unconstitutional action by government agencies. It's even worse when public officials are blatantly lying in ways that contradict dozens of pieces of video evidence. 519825
Samuel Müller @sammuller.bsky.social · 08/07/2025Compute is increasing much faster than data. How can we improve classical supervised learning long term (the underlying tech of most of GenAI)? Our ICML position paper's answer: simply train on a bunch of artificial data (noise) and only do inference on real-world data! 1/n 193
Samuel Müller @sammuller.bsky.social · 28/04/2025I am so proud to co-organize the workshop on foundation models for structured data at ICML. At this workshop, we will discuss on how to further extend the GenAI revolution to tabular data, time series forecasting etc. Consider submitting your work by May 19! icml-structured-fm-workshop.github.ioicml-structured-fm-workshop.github.io Foundation Models for Structured Data 040
Samuel Müller @sammuller.bsky.social · 25/04/2025Could it be that @fchollet.bsky.social is not Francois Chollet?? They have a lot of ML followers 😅 210
Samuel Müller @sammuller.bsky.social · 24/02/2025I believe lmarena.ai scores are not to be trusted, as the people voting are likely to come from the AI labs in the leaderboard and push their own models unintentionally. A thread 🧵lmarena.aiChatbot Arena (formerly LMSYS): Free AI Chat to Compare & Test Best AI Chatbots 111
Samuel Müller @sammuller.bsky.social · 23/02/2025How wrong do you think are the lmarena scores? Grok must be very easy to distinguish from other models in a blind evaluation 000
Samuel Müller @sammuller.bsky.social · 15/01/2025MiniMax-01 takeaways - 7 of 8 layers are linear att - implemented a flash-variant of linear attention + ring-att - post-norm is back in large models! (using deepnorm) - prob. wrong scaling laws, as lr schedule is not adapted (see Chinchilla) Let's see how it fares in the arena! 000
Reposted by Samuel MüllerVíctor @victorbcn.bsky.social · 09/01/2025Los modelos preentrenados para datos tabulares (TabPFN) podrían ser el nuevo state of the art para regresión y clasificación. 🙄 Habrá que probarlo. El GitHub al final del hilo. Éste en concreto es enorme y si se comporta como dicen es un gran salto adelante en el state of the art del campo. 041
Reposted by Samuel Müllerleogrin.bsky.social @leogrin.bsky.social · 09/01/2025Groundbreaking work, congrats to the team!! 🎉 When I started my PhD 3 years ago, our tabular benchmark showed tree-based models miles ahead of neural networks. On the same benchmark, TabPFN v2 now reaches in 10s what CatBoost achieves in 4h of tuning 🤯 141
Reposted by Samuel MüllerLennart Purucker @lennartpurucker.bsky.social · 09/01/2025The tabular foundation model TabPFN v2 is finally public 🎉🥳 This is excellent news for (small) tabular ML! Checkout our Nature article (nature.com/articles/s41...) and code (github.com/PriorLabs/Ta...) 0111
Samuel Müller @sammuller.bsky.social · 08/01/2025This might be the first time after 10 years that boosted trees are not the best default choice when working with data in tables. Instead a pre-trained neural network is, the new TabPFN, as we just published in Nature 🎉 33714
Samuel Müller @sammuller.bsky.social · 04/12/2024While ICLR forced everyone to review, it also had the worst reviewer-paper matching of any conference for me. 2 of the 3 papers were on topics that I never worked on before... :( 030