optimization

CT No.245: Billy Joel mathletics

CT No.244: Explicating "attention" in LLMs

The words we publish and hold up for peer review remain the best representation of our brains at work in the digital world. A published paper is the best way to look closely at the foundational assumptions of LLMs. And those begin with pop culture.

CT No.243: Disambiguation, sliding doors, hallucinations, and madeleines

Transformers take static vector embeddings, which assign single values to every token, and expand their context, nearly simultaneously as they process the context of every other word in the sentence. But who cares, let's listen to a pop song!

CT No.242: What's your vector, Victor?

How to understand tokens and vector embeddings, for word people.

CT No.241: TF? ID(GA)F: The old faithfuls of language processing

Even in the face of "black box" algorithms, the history of artificial intelligence—natural language processing, more specifically—has left plenty of clues. While we can't understand the full equation, we can see how building blocks create common patterns in how current algorithms process language.