RESEARCH

OPEN SOURCE

Dynatask, a new paradigm of AI benchmarking is now available for the AI community

Today, we’re unlocking Dynabench’s full capabilities for the AI community — AI researchers can now create their own custom tasks to better evaluate the performance of NLP models in more.

September 21, 2021

RESEARCH

OPEN SOURCE

Dynatask, a new paradigm of AI benchmarking is now available for the AI community

Today, we’re unlocking Dynabench’s full capabilities for the AI community — AI researchers can now create their own custom tasks to better evaluate the performance of NLP models in more.

September 21, 2021

RESEARCH

New Captum version features more ways to build AI responsibly

September 21, 2021

RESEARCH

NLP

Textless NLP: Generating expressive speech from raw audio

September 09, 2021

Blog Posts

Blog

July 16, 2021

RESEARCH

Blender Bot 2.0: An open source chatbot that builds long-term memory and searches the internet

We’ve built and open-sourced BlenderBot 2.0, the first chatbot that can store and continually access long-term memory, search the internet for timely information, and have sophisticated conversations on nearly any topic. It’s a significant advancement in conversational AI.

July 16, 2021

Learn More

Blog

July 14, 2021

RESEARCH

COMPUTER VISION

Improved supernet training for efficient neural architecture search

We’ve developed two new methods to significantly improve the accuracy of so-called supernets, which have emerged as a powerful way to make network architecture search more efficient. AttentiveNAS and AlphaNet deliver state-of-the-art results on the ImageNet data set.

July 14, 2021

Learn More

Blog

July 07, 2021

RESEARCH

COMPUTER VISION

Audiovisual self-supervised representation learning

We’re sharing new research on using the natural association between video & sound to teach machines to better understand the world. Our self-supervised approach, which is a #CVPR21 best paper candidate, learns directly from sounds & images in videos.

July 07, 2021

Learn More

Blog

July 07, 2021

RESEARCH

COMPUTER VISION

Demystifying a key self-supervised learning technique: Non-contrastive learning

We’re sharing a new theory that attempts to explain one of the mysteries of deep learning: why so-called non-contrastive self-supervised learning often works well.

July 07, 2021

Learn More

Blog

July 06, 2021

RESEARCH

COMPUTER VISION

Better computer vision models by combining Transformers and convolutional neural networks

We’ve developed a new computer vision model called ConVit, which combines two widely used AI architectures — convolutional neural networks (CNNs) and Transformer-based models — in order to overcome some important limitations of each approach on its own.

July 06, 2021

Learn More

Blog

July 01, 2021

RESEARCH

NLP

Democratizing conversational AI systems through new data sets and research

We’re sharing new research and data sets intended to help AI researchers build better conversational AI systems for people around the globe. Our method uses 10x less training data to create conversational AI that can perform unfamiliar, complex tasks.

July 01, 2021

Learn More

Blog

June 30, 2021

RESEARCH

COMPUTER VISION

Habitat 2.0: Training home assistant robots with faster simulation and new benchmarks

Habitat 2.0 will help researchers tackle the next frontier in embodied AI – enabling home assistive robots to be trained in interactive environments and at much faster speeds than its predecessor.

June 30, 2021

Learn More

Blog

June 30, 2021

RESEARCH

ML APPLICATIONS

Using AI to help health experts address the COVID-19 pandemic

Facebook AI is now open-sourcing our entire stack of COVID-19 forecasting models so that response teams, governments, and researchers can use them to further help their communities.

June 30, 2021

Learn More

Blog

June 30, 2021

OPEN SOURCE

Introducing the Habitat-Matterport 3D research data set for training embodied AI

Data sets used to train embodied AI have been in short supply. Now, Facebook AI is partnering with Matterport to make available to the academic research community Matterport’s open-sourced data set of indoor 3D scans.

June 30, 2021

Learn More

Blog

June 29, 2021

RESEARCH

COMPUTER VISION

CVPR 2021 awards given to Facebook AI researchers

We’d like to recognize the Facebook AI researchers who have just received awards at the Conference on Computer Vision and Pattern Recognition (CVPR): Alex Berg, Tamara Berg, Xinlei Chen, and Georgia Gkioxari.

June 29, 2021

Learn More

Blog

June 23, 2021

RESEARCH

COMPUTER VISION

Q&A with Georgia Gkioxari, winner of the 2021 PAMI Young Researcher Award

Facebook AI’s Georgia Gkioxari, who has received the 2021 Pattern Analysis and Machine Intelligence (PAMI) Young Researcher Award, discusses her work in computer vision.