RT @Pavel_Izmailov: I am recruiting Ph.D. students for my new lab at @nyuniversity! Please apply, if you want to work with me on reasoning,….

Recently I've been playing around with a quarter-order-of-magnitude system for simple calculations. It gives better precision than single sig-fig calculations using only four, very intuitive, symbols.

If you miss the NYTimes needle, especially one that is statistically uniform (, you can use this page: I whipped together to reason about the correlations between the swing states tonight as results come in.

Why don't we measure probabilities in degrees?.

In which I try to make sense of most of machine learning:

RT @blester125: Is Kevin onto something? We found that LLMs can struggle to understand compressed text, unless you do some specific tricks.….

RT @noahconst: Ever wonder why we don’t train LLMs over highly compressed text? Turns out it’s hard to make it work. Check out our paper fo….

Each delivery service should use its own distinctive knock.

PaLM 540 Billion, Google's large language model used 4.2 moles of flops to train. 4.2 Moles!.

RT @poolio: Happy to announce DreamFusion, our new method for Text-to-3D!. We optimize a NeRF from scratch using a….

RT @alemi: @dpkingma @poolio To accompany the colab, I've also written a blog post attempting to make sense of the….

RT @dpkingma: Want to understand and/or play with variational diffusion models? . - See for a simple stand-alone im….

RT @ziv_ravid: A pretty cool paper (and I also hope useful) on using pre-training models to create highly informative priors for downstream….

RT @ethansdyer: 1/ Super excited to introduce #Minerva 🦉(. Minerva was trained on math and science found on the web….

RT @Chitwan_Saharia: We are thrilled to announce Imagen, a text-to-image model with unprecedented photorealism and deep language understand….

you can verify with `echo -n "answer" | md5sum`.

here are the next few days wordle answers as md5 hashes .2022-01-11 = 0b18a3d7b9c43ff1750d2baa4606b8d0.2022-01-12 = 047fb90408a79f189d51cbcea168b1a5.2022-01-13 = ab3358313efb03210a1babfb372246f1.2022-01-14 = d821e448212defd91ac1e67f9653a34d.

RT @samscub: We are presenting our paper "Does Knowledge Distillation Really Work?" at #NeurIPS2021 poster session 2 today - come check it….

RT @venkvis: Excited to kick-start focus #SciML series on #ML meets Info theory and statistical mechanics! Amazing speaker/session chair li….

RT @polkirichenko: While most papers on knowledge distillation focus on student accuracy, we investigate the agreement between teacher and….