Machine learning

Three years ago, Google’s AlphaFold pulled off the biggest artificial intelligence breakthrough in science to date, accelerating molecular research and kindling deep questions about why we do science.

Researchers are drawing on ideas from game theory to improve large language models and make them more consistent.

Google DeepMind’s AlphaFold3 and other deep learning algorithms can now predict the shapes of interacting complexes of protein, DNA, RNA and other molecules, better capturing cells’ biological landscapes.

The computer scientist Ellie Pavlick is translating philosophical concepts such as “meaning” into concrete, testable ideas.

By apparently overtraining them, researchers have seen neural networks discover novel solutions to problems.

Large language models do better at solving problems when they show their work. Researchers are beginning to understand why.

Erasing key information during training results in machine learning models that can learn new languages faster and more easily.

A new study suggests that so-called emergent abilities actually develop gradually and predictably, depending on how you measure them.

Far from being “stochastic parrots,” the biggest large language models seem to learn enough skills to understand the words they’re processing.