AI Technology & Industry Review
Google recently introduced mT5, a multilingual variant of its “Text-to-Text Transfer Transformer” (T5), pretrained on a new Common Crawl-based dataset covering 101 languages.
“Trust in AI systems is becoming, if not already, the biggest barrier for enterprises — as they start to move from exploring AI or potentially piloting or doing some proof of concept works into deploying AI into a production system”
ICLR 2021 submission proposes LambdaNetworks, a transformer-specific method that reduces costs of modeling long-range interactions for CV and other applications.
Facebook AI open-sourced a multilingual machine translation (MMT) model that translates between any pair of 100 languages without relying on English data.
A new Facebook AI and CMU renewable energy storage project could enable labs to perform days of electrocatalyst screening and calculations in just seconds.
“The Computational Limits of Deep Learning” first author Neil Thompson of MIT says DL’s economic and environmental footprints are growing worrying fast
Pinkney and Adler NeurIPS 2020 workshop paper enables realistic image generation in domains such as animation and ukiyo-e with creative control on the output.
ICLR 2021 submitted paper proposes efficient VAEs that outperform PixelCNN-based autoregressive models in log-likelihood on natural image benchmarks.
ByteDance introduces a high-resolution piano transcription system trained by regressing the precise onset and offset times of piano notes and pedals.
PwC and arXiv jointly announced their partnership yesterday, unveiling a convenient new Code tab on the abstract page of arXiv Machine Learning articles.
ICLR 2021 paper An Image Is Worth 16×16 Words: Transformers for Image Recognition at Scale suggests Transformers can outperform top CNNs on CV at scale.
NeurIPS 2020 released its list of accepted papers this week with Google, Stanford, and MIT as the top affiliations.
NVIDIA, Mass General Brigham and 20 global hospitals launch federated learning initiative EXAM to build AI model for COVID-19 patient oxygen need prediction
Google AI researchers developed a sign language detection model for video conferencing applications that can perform real-time identification of a person signing as an active speaker.
A team from Google, University of Cambridge, DeepMind, and Alan Turing Institute have proposed a new type of Transformer dubbed Performer, based on a Fast Attention Via positive Orthogonal Random features (FAVOR+) backbone mechanism.
Novel model uses a quality estimator and evolutionary optimization to search the latent space of GANs trained on limited datasets.
Researchers introduced retrieval-augmented generation - a hybrid, end-to-end differentiable model that combines an information retrieval component with a seq2seq generator.
Facebook AI researchers have open-sourced the new wav2vec 2.0 algorithm for self-supervised language learning.
UIUC, Adobe Research and University of Oregon propose HDMatt, a Deep Learning-based image matting Cross-Patch Context module for high-resolution image inputs.
Augmented Temporal Contrast (ATC), a new unsupervised learning (UL) task for learning visual representations agnostic to rewards and without degrading the control policy.
From an augmented view of an image, the researchers trained the online network to predict the target network representation of the same image under a different augmented view.
OpenAI sets out to advance methods for training large-scale language models on objectives that more closely capture human preferences.
This research addresses a well-known phenomenon regarding large batch sizes during training and the generalization gap.
OpenAI researchers introduce GPT-f, an automated prover and proof assistant for the Metamath formalization language.
Novel attention condensers designed to enable the building of low-footprint, highly-efficient deep neural networks for on-device speech recognition on the edge.
Researchers introduce a test covering topics such as elementary mathematics, designed to measure language models’ multitask accuracy.
Researchers introduced a novel flow-based video completion algorithm that compares favourably with the state-of-the-art in the field.
This research proposes an efficient and cost-effective solution for multi-frame video interpolation.
The ReDNA Labs research team has devised a new lightweight approach called IGLOO which allows to deal with sequences up to 25,000 steps long.
Although OpenAI hasn’t yet officially announced the GPT-3 pricing scheme, Branwen’s sneak peek has piqued the interest of the NLP community.
“Wav2Lip,” a novel lip-synchronization model that outperforms current approaches by a large margin in both quantitative metrics and human evaluations.
Facebook AI this week released a new high-speed library called Opacus.
Jingxiong highway, a highway with designated lanes for unmanned vehicles, is expected to be completed by the end of this year and open to traffic in June 2021.
On August 28, China’s Ministry of Commerce and Ministry of Science and Technology released the new updated version of the Catalogue of Technologies Prohibited and Restricted from Export.
The paper outlines a predictive model we’ve developed that has the potential to help significantly reduce wasteful healthcare spending.
Researchers propose reducing the workloads of urban planners by introducing deep learning systems to handle some of their responsibilities.
Intel Labs researchers have proposed a novel method for building a robot called “OpenBot” on just a US$50 budget.
Researchers proposed a new model that is designed to spot deepfakes by looking at subtle visual artifacts.
A novel volumetric capture system that is capable of fully capturing clothed human bodies in real-time using only a single RGB webcam.
The 16th European Conference on Computer Vision (ECCV) kicked off on Sunday as a fully online conference. In the Conference Opening Session this morning, the ECCV organizing committee announced the conference’s paper submission stats and Best Paper selections.