Skip to main content

The latest research from Google

Distilling step-by-step: Outperforming larger language models with less training data and smaller model sizes

Large language models (LLMs) have enabled a new data-efficient learning paradigm wherein they can be used to solve unseen new tasks via zero-shot or few-shot prompting. However, LLMs are challenging to deploy for real-world applications due to their sheer size. For instance, serving a single 175 billion LLM requires at least 350GB of GPU memory using specialized infrastructure, not to mention that today's state-of-the-art LLMs are composed of over 500 billion parameters. Such computational requirements are inaccessible for many research teams, especially for applications that require low latency performance.

MediaPipe FaceStylizer: On-device real-time few-shot face stylization

On-device content distillation with graph neural networks

World scale inverse reinforcement learning in Google Maps

Differentially private median and more

A novel computational fluid dynamics framework for turbulent flow research

TSMixer: An all-MLP architecture for time series forecasting

WeatherBench 2: A benchmark for the next generation of data-driven weather models

Modeling and improving text stability in live captions

SayTap: Language to quadrupedal locomotion

RO-ViT: Region-aware pre-training for open-vocabulary object detection with vision transformers