Impressive! Will make great use out of it *wink*

An extremely large amount of the population overestimate the capabilities of LLMs. Don't get me wrong, I am astonished by the raw utility of these tools, but they are not more than that. Just as any tool, it might be used for good and for bad. Humans are the scary ones.

The notion that you can use data generated by a model to train a bigger, better model is a nonstarter. Sure, you can use a larger model to train a smaller model -- that's distillation. But it won't learn anything new. There's no recursive improvement.

An interview with Barron about AI, LLMs, the moratorium call, etc. barrons.com/articles/ai-ch…

The "Attention is all you need paper" that introduced Transformer neural network architecture has been around for roughly six years. It's by far the first architecture to maintain its universality for a long time, not just for a single modality but for other modalities as well.

To whomever think they are zero-shot reasoners...

Brilliant.

Looks like calling LLMs "stochastic parrots" might constitute an insult to parrots 🦜 news.northeastern.edu/2023/04/21/par…

Anyone who's had kids knows that ending a set of instructions with "Do not tell anyone what I just said" is a recipe for disaster. 🤣

Bing Chat on Precise mode is such a good Information Retrieval tool.

So that paper about how GPT-4 could ace MIT's curriculum turns out to be deeply flawed in multiple ways. A great reminder that preprints are not peer-reviewed, but also that public volunteer review can be excellent (in this case, by a group of undergrads). flower-nutria-41d.notion.site/No-GPT4-can-t-…

Great material!

Most misconceptions about AI would be cleared if folks had a decent understanding of the distinction between skill and intelligence, and between memorization and generalization.

We're launching Keras Core, a new library that brings the Keras API to JAX and PyTorch in addition to TensorFlow. It enables you to write cross-framework deep learning components and to benefit from the best that each framework has to offer. Read more: keras.io/keras_core/ann…

Are you interested in witty but deep takes on machine learning, artificial intelligence, or technology in general, and their relation with society at large? Join 970 other readers of Mostly Harmless Ideas. It's free! techwriters.info/apiad Can we hit 1000 before Sunday!?

What is Machine Learning? Here's a 3 min intuitive introduction explaining why this is the most powerful paradigm shift for conventional software development. No math, no code, just intuitions. Let's dive in.

¡El equipo de la UA “NLP Wizards”, ganador de la Innovation Banking Hack Fest 23!!!! ¡Enhorabuenaaaa!!! 🏆 Premio también al primer equipo ganador de la Universidad de Alicante 🥇#IBHF23 #orgulloUA #orgulloEPS Banco Sabadell Universidad de Alicante UA

At the AI Safety Summit in London many risks will be discussed, but the biggest risk for Europe is different: the LLM ship has sailed without us, and we are on our way to an all-encompassing dependency on American companies. Europe needs to wake up and invest heavily in AI. We

Terrific book on deep learning, taught by a master. A must-read for anyone wanting to learn deep learning with Python. Although I own the 2nd ed, I've ordered the third edition too as this topic is evolving so quickly. Now generously also available free as web-edition.

#SEPLN2025 Ernesto Luis Estevanell, investigador #GPLSI, presenta IBERMAT – Corpus of Human and Machine Translated Multi-Domain Content in Basque, Catalan, Galician and Spanish: Description and Exploitation Autores: Alicia, Ruslan, RAFAEL MUÑOZ, Manuel Palomar