Our plan to host the 2022 Meta Research Testing and Verification (TAV) Symposium in person has been postponed because of COVID-related event restrictions.Read More
EgoObjects: A large-scale egocentric dataset for object understanding
In this blog, we’ll introduce the EgoObjects pilot version, consisting of 9,273 videos (30+ hours) for 368 categories of objects, with a total of 654K annotations.Read More
Four analytics best practices we adopted — and why you should too
Data-driven decisions are everywhere. Examples include launching a new feature in your favorite app, ensuring that the products do not discriminate against certain groups…Read More
Introducing LLaMA: A foundational, 65-billion-parameter large language model
Introducing LLaMA (Large Language Model Meta AI), a state-of-the-art foundational large language model.Read More
Token Merging: Your ViT but faster
Meta AI is sharing new research to reduce the latency of existing Vision Transformer (ViT) models without the need for additional training. Our approach, called Token Merging…Read More
The future of time-series forecasting, with RFP winner B. Aditya Prakash
From time to time, Meta invites academics to propose research in specific areas that align with our mission of building community and bringing the world closer together.Read More
Delegated Multi-key Private Matching for Compute: Improving match rates and enabling adoption
The ever-increasing adoption of privacy-enhancing technologies (PETs) provides new layers of privacy in areas such as secure data analytics and machine learning. The first step…Read More
How Meta uses AI to better understand people’s ages on our platforms
Providing age-appropriate experiences for the billions of people who use our services around the world is an important element of what we do.Read More
Atlas: Few-shot learning with retrieval augmented language models
We released the code for our Atlas project [1] on GitHub, as well as pretrained Atlas model checkpoints, an index, and Wikipedia corpora. We present how to build a Q&A system…Read More
The future of time-series forecasting, with RFP winner B. Aditya Prakash
From time to time, Meta invites academics to propose research in specific areas that align with our mission of building community and bringing the world closer together.Read More