How do LLMs pick the next word? They don’t choose words directly: they only output word probabilities. 📊 Greedy decoding, top-k, top-p, min-p are methods that turn these probabilities into actual text. In this video, we break down each method and show how the same model can

@growing_daniel unlike claude code, junior devs learn from their mistakes instead of having to be corrected for the same things over and over and over again until the sun burns out

"A stylish woman in a beige coat walking confidently past a moving train at a modern train station." Create images and videos in seconds with Grok Imagine.

Can an AI model predict perfectly and still have a terrible world model? What would that even mean? Our new ICML paper formalizes these questions One result tells the story: A transformer trained on 10M solar systems nails planetary orbits. But it botches gravitational laws 🧵

Thanks for sharing! 🙂Here another abstract for the visual people. 😀

This is really BAD news of LLM's coding skill. ☹️ The best Frontier LLM models achieve 0% on hard real-life Programming Contest problems, domains where expert humans still excel. LiveCodeBench Pro, a benchmark composed of problems from Codeforces, ICPC, and IOI (“International

Ever wondered how linear RNNs like #mLSTM (#xLSTM) or #Mamba can be extended to multiple dimensions? Check out "pLSTM: parallelizable Linear Source Transition Mark networks". #pLSTM works on sequences, images, (directed acyclic) graphs. Paper link: https://t.co/nU7626uHWK

After ~6 years of building these types of architectures (starting with BERT, eg see Baleen), I think calling these multi-agent systems is a distraction. This is just software. Happens to be AI software. It doesn’t seem so complicated once you internalize it’s just a program.

Bro how was the show Silicon Valley so consistently 10 years ahead of its time 🤣

https://t.co/eOp5DbkXd7

https://t.co/Z4vRHmjQoc

The data science revolution is getting closer. TabPFN v2 is published in Nature: https://t.co/Ybb15pnZ5P On tabular classification with up to 10k data points & 500 features, in 2.8s TabPFN on average outperforms all other methods, even when tuning them for up to 4 hours🧵1/19

I don't think people really appreciate how simple ARC-AGI-1 was, and what solving it really means. It was designed as the simplest, most basic assessment of fluid intelligence possible. Failure to pass signifies a near-total inability to adapt or problem-solve in unfamiliar

How Nobel Prizes will be given in the future 😑 #Physics #NobelPrize #NobelPrize2024 #scicomm

Making a video with @plazmapunk is fun. I never know what I will get to see until the video is generated. Music made with @mytracksy

𝐁𝐌𝟒𝟐 - 𝐓𝐡𝐞 𝐌𝐢𝐬𝐬𝐢𝐧𝐠 𝐁𝐞𝐧𝐜𝐡𝐦𝐚𝐫𝐤 Qdrant released this week an interesting new approach that claims to replace BM25/lexical search. They just sadly forgot to do proper benchmarking. As it turns out: BM42 is way worse than BM25.

the delorean is a conference room THE DELOREAN IS A CONFERENCE ROOM

We are excited to share now the code for xLSTM! 🚀 Happy to hear about your first experiments with it! 🧪 https://t.co/FC3hEz8kFZ

9 out of 10 RAG applications would be better off by establishing a simple baseline before introducing text embeddings.

🔬🎙️ User Research and Interactive #recsys with @McWillemsen from @jadatascience Episode #21 of #recsperts is out! We discuss how users exert control over #recommendations, decision #psychology and user-centric evaluation - go check it out: https://t.co/Rmq0xyZcq0

Ivan Herreros at #CAIML: convex optimization and text-embeddings @CologneAIML #AI #ML #Ki