With recent improvements in natural language generation (NLG) models for various applications, it has become imperative to have the means to identify and evaluate whether NLG output is only sharing verifiable information about the external world. In this work, we present a new evaluation framework entitled Attributable to Identified Sources (AIS) for assessing the output of natural language generation models, when such output pertains to the external world. We first define AIS and introduce a two-stage annotation pipeline for allowing annotators to appropriately evaluate model output according to AIS guidelines. We empirically validate this approach on generation datasets spanning three tasks (two conversational QA datasets, a summarization dataset, and a table-to-text dataset) via human evaluation studies that suggest that AIS could serve as a common framework for measuring whether model-generated statements are supported by underlying sources. We release guidelines for the human evaluation studies.

Link prediction is an important task that has wide applications in various domains. However, the majority of existing link prediction approaches assume the given graph follows homophily assumption, and designs similarity-based heuristics or representation learning approaches to predict links. However, many real-world graphs are heterophilic graphs, where the homophily assumption does not hold, which challenges existing link prediction methods. Generally, in heterophilic graphs, there are many latent factors causing the link formation, and two linked nodes tend to be similar in one or two factors but might be dissimilar in other factors, leading to low overall similarity. Thus, one way is to learn disentangled representation for each node with each vector capturing the latent representation of a node on one factor, which paves a way to model the link formation in heterophilic graphs, resulting in better node representation learning and link prediction performance. However, the work on this is rather limited. Therefore, in this paper, we study a novel problem of exploring disentangled representation learning for link prediction on heterophilic graphs. We propose a novel framework DisenLink which can learn disentangled representations by modeling the link formation and perform factor-aware message-passing to facilitate link prediction. Extensive experiments on 13 real-world datasets demonstrate the effectiveness of DisenLink for link prediction on both heterophilic and hemophiliac graphs. Our codes are available at https://github.com/sjz5202/DisenLink

Nonnegative Tucker decomposition (NTD), a tensor decomposition model, has received increased interest in the recent years because of its ability to blindly extract meaningful patterns, in particular in Music Information Retrieval. Nevertheless, existing algorithms to compute NTD are mostly designed for the Euclidean loss. This work proposes a multiplicative updates algorithm to compute NTD with the beta-divergence loss, often considered a better loss for audio processing. We notably show how to implement efficiently the multiplicative rules using tensor algebra. Finally, we show on a music structure analysis task that unsupervised NTD fitted with beta-divergence loss outperforms earlier results obtained with the Euclidean loss.

NEURODATA is a global technology company dedicated to creating and scaling AI solutions by leveraging deep expertise and a common machine learning platform. Through meaningful engagement and collaboration, we partner with companies to build and operationalize impactful end-to-end AI solutions across a wide variety of industries and use cases. Our experience in computer vision, NLP, and recommendation engines in real world scenarios enables us to deliver solutions that integrate seamlessly with customer processes to solve the most complex business challenges. Our team consists of expert data scientists, software engineers, and AI business leaders in Tunisia and Montreal, Canada. As the Center of Excellence for AI embedded, we place the highest priority on the ethical adoption, development, and delivery of AI technologies and solutions, with the goal of better business outcomes that usher in a more resilient and inclusive world.

Ayman Al Zawahiri, the terrorist killed in a U.S. drone strike in Afghanistan Monday, was a top deputy to al Qaeda leader Usama bin Laden before taking the helm of the organization after his predecessor's death in 2011. A drone strike on a Kabul home took him out over the weekend, Fox News reported earlier. Taliban spokesman Zabihullah Mujahid confirmed and condemned the attack on Twitter, calling it "a clear violation of international principles," according to a translation of the thread. However, the 2020 Doha Agreement, which preceded the Biden administration's highly criticized withdrawal of U.S. troops from Afghanistan last year, called for the Taliban to combat terrorism within the country. Al Zawahiri was also a doctor and founder of the Egyptian Islamic Jihad terror group, which later merged with al-Qaeda, according to authorities.

CloudFactory, a global leader in human-in-the-loop artificial intelligence (AI), announced that Pieter Nel has joined as Chief Technology Officer (CTO). Nel brings more than 20 years of experience, across three continents, in technology strategy and software engineering management at fast-growth technology companies. As CTO, he will lead the technology and machine learning (ML) teams, continuously evolving CloudFactory's platform as a key enabler for clients' successful AI deployments. "Considering all successful AI deployments include humans in the loop, CloudFactory is positioned perfectly to support clients with our experienced annotation workforce and building the infrastructure to enable human-in-the-loop AI deployments." Nel previously served as CTO at Ocrolus, where he scaled the New York company's human-in-the-loop AI document processing product.

For millennia, we've been told that asking questions was the path to enlightenment. But in the surveillance age, it might land you in jail. That's the danger of a new search tactic that police are increasingly turning to in their constant campaign to transform our phones and devices into evidence against us: keyword warrants. One Denver court may soon rule on whether they can continue as a policing tactic--and in the post-Roe era, the wrong decision could put abortion seekers in unprecedented danger. Police have used web browser history and search engine data in their investigations for about as long as the data has existed, but keyword warrants are different--a digital dragnet to find every user who searches for a specific person, place or thing.

Over the weekend, debris from an out-of-control Chinese rocket crashed to Earth over the Indian and Pacific oceans. There had been fears that pieces of the 23-tonne Long March 5B booster could come down over a populated area, but experts had said the probability of this was extremely low. Nevertheless, NASA hit out at China by accusing Beijing of not sharing the'specific trajectory information' needed to calculate where possible debris might fall. Elsewhere at the weekend, a 10ft (3m) piece of space junk – thought to be from one of Elon Musk's spacecrafts – crashed into a farmer's property in Australia at around 15,500mph (25,000km/h). The object, believed to be part of the SpaceX Crew-1 craft, was found in a sheep paddock by a farmer living on a large property in the Snowy Mountains in New South Wales.

There's nothing like migration to reveal how things that seem natural may be artifacts of culture. When I left India for college in England, I was surprised to find that pinching my Adam's apple didn't mean, as I had thought it meant everywhere, "on my honor." I learned to expect only mockery at the side-to-side tilts of the head with which I expressed degrees of agreement or disagreement, and trained myself to keep to the Aristotelian binary of nod and shake. Around that time, I also learned--from watching the British version of "The Office"--that the word "cringe" could be an adjective, as in the phrase "so cringe." It turned out that there was a German word for the feeling inspired by David Brent, the cringe-making boss played by Ricky Gervais in the show: Fremdschämen--the embarrassment one feels when other people have, perhaps obliviously, embarrassed themselves.

Health monitoring applications increasingly rely on machine learning techniques to learn end-user physiological and behavioral patterns in everyday settings. Considering the significant role of wearable devices in monitoring human body parameters, on-device learning can be utilized to build personalized models for behavioral and physiological patterns, and provide data privacy for users at the same time. However, resource constraints on most of these wearable devices prevent the ability to perform online learning on them. To address this issue, it is required to rethink the machine learning models from the algorithmic perspective to be suitable to run on wearable devices. Hyperdimensional computing (HDC) offers a well-suited on-device learning solution for resource-constrained devices and provides support for privacy-preserving personalization. Our HDC-based method offers flexibility, high efficiency, resilience, and performance while enabling on-device personalization and privacy protection. We evaluate the efficacy of our approach using three case studies and show that our system improves the energy efficiency of training by up to $45.8\times$ compared with the state-of-the-art Deep Neural Network (DNN) algorithms while offering a comparable accuracy.