Neural networks

Larger models can pull off greater feats, but the accessibility and efficiency of smaller models make them attractive tools.

By training machine learning models with enough examples of basic science, Miles Cranmer hopes to push the pace of scientific discovery forward.

Recent results show that large language models struggle with compositional tasks, suggesting a hard limit to their abilities.

Certain grammatical rules never appear in any known language. By constructing artificial languages that have these rules, linguists can use neural networks to explore how people learn.

Researchers got a better look at the thoughts of chatbots, amateurs learned exactly how complicated simple systems can be, and quantum computers passed an essential milestone.

How do we know if a large language model is lying? Letting AI systems argue with each other may help expose the truth.

To better understand machine learning algorithms, Lenka Zdeborová treats them like physical materials.

Machines work with words by embedding their relationships with other words in a string of numbers.

By tapping into a decades-old mathematical principle, researchers are hoping that Kolmogorov-Arnold networks will facilitate scientific discovery.