InfoQ Homepage Facebook Content on InfoQ
-
Meta Open-Sources 175 Billion Parameter AI Language Model OPT
Meta AI Research released Open Pre-trained Transformer (OPT-175B), a 175B parameter AI language model. The model was trained on a dataset containing 180B tokens and exhibits performance comparable with GPT-3, while only requiring 1/7th GPT-3's training carbon footprint.
-
Meta Announces Conversational AI Model Project CAIRaoke
Meta AI Research recently announced Project CAIRaoke, an end-to-end deep-learning model for digital assistants. Project CAIRaoke is currently being used in Meta's Portal device and outperforms a previous conversational model when evaluated on a reminder task.
-
Infer New Deadlock Detector for Android Aims to Be Efficient for Large Codebases
In a joint research, London University College and Facebook researchers created a new deadlock detector for Android Java code now available as part of open-source Infer static analysis tool. The new analyzer is able to process large codebases efficiently thanks to its incremental approach specifically designed for integration in a CI pipeline.
-
Meta Open-Sources Multi-Modal AI Algorithm Data2vec
Meta AI recently open-sourced data2vec, a unified framework for self-supervised deep learning on images, text, and speech audio data. When evaluated on common benchmarks, models trained using data2vec perform as well as or better than state-of-the-art models trained with modality-specific objectives.
-
Meta Unveils AI Supercomputer for the Metaverse
Meta has unveiled its AI Research SuperCluster (RSC) supercomputer, aimed at accelerating AI research and helping the company build the metaverse. The RSC will help the company build new and better AI models, working across hundreds of different languages, and to develop new augmented reality tools.
-
Austrian DPA Ruling against Google Analytics Paves the Way to EU-based Cloud Services
In a recent ruling, the Austrian data regulator declared the use of Google Analytics unlawful based on EU GDPR regulation. While the ruling is very specifically argued and worded, its implications go well beyond this particular case.
-
Meta and AWS to Collaborate on PyTorch Adoption
Meta and AWS will work together to improve the performance for customers of applications running PyTorch on AWS and accelerate how developers build, train, deploy, and operate artificial intelligence and machine-learning models.
-
Facebook Open-Sources Two Billion Parameter Multilingual Speech Recognition Model XLS-R
Facebook AI Research (FAIR) open-sourced XLS-R, a cross-lingual speech recognition (SR) AI model. XSLR is trained on 436K hours of speech audio from 128 languages, an order of magnitude more than the largest previous models, and outperforms the current state-of-the-art on several downstream SR and translation tasks.
-
Improve Android App Startup Times Like Facebook and Google Do
In a recent article, engineers from Google and Facebook provided their hints at what matters the most when reducing the time it takes for an Android app to become responsive on launch and ensuring an optimal user experience.
-
Facebook Open-Sources GHN-2 AI for Fast Initialization of Deep-Learning Models
A team from Facebook AI Research (FAIR) and the University of Guelph have open-sourced an improved Graph HyperNetworks (GHN-2) meta-model that predicts initial parameters for deep-learning neural networks. GHN-2 executes in less than a second on a CPU and predicts values for computer vision (CV) networks that achieve up to 77% top-1 accuracy on CIFAR-10 with no additional training.
-
Building Large-Scale iOS Apps at Airbnb
The Airbnb iOS team addressed the challenge of its growing mobile app codebase and complexity by adopting new tools and processes, including a modern build systems, module types, and dev apps.
-
Facebook CSS-in-JS Solution Stylex Introduced at React Finland 2021
Naman Goel, software engineer at Facebook, recently presented Stylex at React Finland 2021. Stylex is Facebook’s custom CSS-in-JS solution used for the new facebook.com website. StyleX alleviates key pain points of CSS-in-JS for large React applications (unused styles, large CSS files, CSS-in-JS library size). Goel anticipates that Stylex will be open-sourced by the end of 2021.
-
Flow Type Checker No Longer Just JavaScript with Types, Centers of Facebook Needs
Vladan Djeric, engineering manager supporting the Flow team at Facebook, announced that the Flow type checker will go beyond being just JavaScript with types and introduce new features based on Facebook’s internal user needs. In particular, Flow strives to deliver fast type-checking on large, complicated codebases. Flow’s vision is in contrast with TypeScript’s — JavaScript with syntax for types.
-
Facebook Open-Sources Computer Vision Model Multiscale Vision Transformers
Facebook AI Research (FAIR) recently open-sourced Multiscale Vision Transformers (MViT), a deep-learning model for computer vision based on the Transformer architecture. MViT contains several internal resolution-reduction stages and outperforms other Transformer vision models while requiring less compute power, achieving new state-of-the-art accuracy on several benchmarks.
-
Facebook Superpack Leverages Code Analysis for Android App Compression
In a recent article, Facebook described its novel technique for Android app compression, Superpack, which combines compiler analysis with data compression. While not yet available for everyone, Facebook is hoping to open source it.