Neural Networks

Baidu debuted two models: a lightweight, open-weights, vision-language model and a giant, proprietary, multimodal model built to take on U.S. competitors.

The approach known as LoRA streamlines fine-tuning by training a small adapter that modifies a pretrained model’s weights at inference. Researchers built a model that generates such adapters directly.

Researchers built a model that integrates satellite imagery and other sensor readings across the entire surface of the Earth to reveal patterns of climate, land use, and other features.

DINOv2 showed that a vision transformer pretrained on unlabeled images could produce embeddings that are useful for a wide variety of tasks. Now it has been updated to improve the performance of its embeddings in segmentation and other vision tasks.

Reducing the number of bits used to represent each parameter in a neural network from, say, 16 bits to 8 bits shrinks the network’s size and boosts its speed. Researchers took this approach to an extreme: They built a competitive large language model whose weights are limited to three values.

Improving a large language model’s factual accuracy typically requires making it bigger, which in turn, involves more computation. Researchers devised an architecture that enables models to recall relevant details without significantly increasing the amount of computation required.

To date, efforts to decode what people are thinking from their brain waves often relied on electrodes implanted in the cortex. New work used devices outside the head to pick up brain signals that enabled an AI system, as a subject typed, to accurately guess what they were typing.

A new generation of robots can handle some household chores with unusual skill.

It’s not necessary to activate all parts of a large language model to process a given input. Using only the necessary parts saves processing.

Researchers used an AI system to identify animal cell types from gene sequences, including a cell type that conventional approaches had discovered only in the past year. 

Pruning weights from a neural network makes it smaller and faster, but it can take a lot of computation to choose weights that can be removed without degrading the network’s performance.

A neural network detected early signs of pancreatic cancer more effectively than doctors who used the usual risk-assessment criteria. Researchers at MIT and oncologists at Beth Israel Medical Center in Boston...

A neural network makes music for unaccompanied vocal tracks. Chris Donahue, Antoine Caillon, Adam Roberts, and colleagues at Google proposed SingSong, a system that generates musical accompaniments for sung melodies. You can listen to its output here.

Biologists used neural networks to find a new class of antibiotics. Researchers at MIT and Harvard trained models to screen chemical compounds for those that kill methicillin-resistant Staphylococcus aureus (MRSA), the deadliest among bacteria that have...

Online sorting algorithms run trillions of times a day to organize lists according to users’ interests. New work found faster alternatives. Daniel J. Mankowitz and colleagues at Google developed AlphaDev, a system that learned to generate algorithms that sort three...