Machine-learning systems can be duped or confounded by situations they haven't seen before. A self-driving car gets flummoxed by a scenario that a human driver could handle easily. An AI system laboriously trained to carry out one task (identifying cats, say) has to be taught all over again to do something else (identifying dogs). In the process, it's liable to lose some of the expertise it had in the original task. Computer scientists call this problem "catastrophic forgetting."

Neural natural language generation (NLG) models have recently shown remarkable progress in fluency and coherence. However, existing studies on neural NLG are primarily focused on surface-level realizations with limited emphasis on logical inference, an important aspect of human thinking and language. In this paper, we suggest a new NLG task where a model is tasked with generating natural language statements that can be \emph{logically entailed} by the facts in an open-domain semi-structured table. To facilitate the study of the proposed logical NLG problem, we use the existing TabFact dataset \cite{chen2019tabfact} featured with a wide range of logical/symbolic inferences as our testbed, and propose new automatic metrics to evaluate the fidelity of generation models w.r.t.\ logical inference. The new task poses challenges to the existing monotonic generation frameworks due to the mismatch between sequence order and logical order. In our experiments, we comprehensively survey different generation architectures (LSTM, Transformer, Pre-Trained LM) trained with different algorithms (RL, Adversarial Training, Coarse-to-Fine) on the dataset and made following observations: 1) Pre-Trained LM can significantly boost both the fluency and logical fidelity metrics, 2) RL and Adversarial Training are trading fluency for fidelity, 3) Coarse-to-Fine generation can help partially alleviate the fidelity issue while maintaining high language fluency. The code and data are available at \url{https://github.com/wenhuchen/LogicNLG}.

Citation parsing, particularly with deep neural networks, suffers from a lack of training data as available datasets typically contain only a few thousand training instances. Manually labelling citation strings is very time-consuming, hence synthetically created training data could be a solution. However, as of now, it is unknown if synthetically created reference-strings are suitable to train machine learning algorithms for citation parsing. To find out, we train Grobid, which uses Conditional Random Fields, with a) human-labelled reference strings from 'real' bibliographies and b) synthetically created reference strings from the GIANT dataset. We find that both synthetic and organic reference strings are equally suited for training Grobid (F1 = 0.74). We additionally find that retraining Grobid has a notable impact on its performance, for both synthetic and real data (+30% in F1). Having as many types of labelled fields as possible during training also improves effectiveness, even if these fields are not available in the evaluation data (+13.5% F1). We conclude that synthetic data is suitable for training (deep) citation parsing models. We further suggest that in future evaluations of reference parsers both evaluation data similar and dissimilar to the training data should be used for more meaningful evaluations.

Semantic Question Answering (QA) is the key technology to facilitate intuitive user access to semantic information stored in knowledge graphs. Whereas most of the existing QA systems and datasets focus on entity-centric questions, very little is known about the performance of these systems in the context of events. As new event-centric knowledge graphs emerge, datasets for such questions gain importance. In this paper we present the Event-QA dataset for answering event-centric questions over knowledge graphs. Event-QA contains 1000 semantic queries and the corresponding English, German and Portuguese verbalisations for EventKG - a recently proposed event-centric knowledge graph with over 970 thousand events.

Our comprehensive guide to how chatbots will develop in 2020 and beyond. Artificial intelligence is the hottest talking point for business users looking to improve their efficiency, deliver new ideas and take the next steps in the transition to a digital enterprise. AI and chatbots are helping democratise business, empower startups and help build new partnerships, something that every organisation needs to prepare for. "Every business is a technology business" was one of the mantras of the decade just concluded. Every company across every vertical and market started working and communicating with smartphones, using cloud services to open up their data and adopted as-a-service solutions to reduce the cost of doing business and broaden their business base and the opportunities for workers. Ten years ago, specialists were needed to manage databases and build websites. Now anyone with a plan can build an entire company out of off-the-shelf parts, sell across the world without leaving their desk. They can pick advice from a huge range of sources to grow the business and partner with a massive range of organisations to deliver whatever they sell. Now as we move into the 2020s, enterprises and startups alike are taking the next step, adopting AI and bringing smart services into their organisations. It has already started with chatbots and analytics tools, but is already expanding to business-enabling technology, using a mix of machine learning, deep learning, computer vision, natural language processing, machine reasoning (MR), and deep or strong AI. Companies will continue to deploy AI for intelligent robotic process automation, computer vision tasks, and machine learning applications.

Text-to-image synthesis is the task of generating images from text descriptions. Image generation, by itself, is a challenging task. When we combine image generation and text, we bring complexity to a new level: we need to combine data from two different modalities. Most of recent works in text-to-image synthesis follow a similar approach when it comes to neural architectures. Due to aforementioned difficulties, plus the inherent difficulty of training GANs at high resolutions, most methods have adopted a multi-stage training strategy. In this paper we shift the architectural paradigm currently used in text-to-image methods and show that an effective neural architecture can achieve state-of-the-art performance using a single stage training with a single generator and a single discriminator. We do so by applying deep residual networks along with a novel sentence interpolation strategy that enables learning a smooth conditional space. Finally, our work points a new direction for text-to-image research, which has not experimented with novel neural architectures recently.

Blue whales swim to the surface to feed on krill as it helps them to conserve energy, according to a new study that involved amazing drone footage of the mammals. Experts from Oregon State University found that feeding on the ocean's surface plays an important role in the hunt for food among New Zealand blue whales. Blue whales are the largest mammals on Earth and have to carefully balance the cost of energy they get from food with the cost of energy used in getting the food. Researchers say the marine mammals forage for krill in areas where they are densely packed and found near the surface of the water to cut their dive time. The Oregon team found that the blue whales do this to conserve on the energetic costs of feeding such as diving, holding their breath or opening their mouths.

It could be argued that scientists create superpowers in their labs. If Aram Galstyan, director of the Artificial Intelligence Division at the USC Viterbi Information Sciences Institute (ISI) had to pick just one superpower, it would be the ability to predict the future. What will be the daily closing price of Japan's Nikkei 225 index at the end of next week? How many 6.0 or stronger earthquakes will occur worldwide next month? Galstyan and a team of researchers at USC ISI are building a system to answer such questions.

The Charisma Machine chronicles the life and legacy of the One Laptop Per Child project and explains why--despite its failures--the same utopian visions that inspired OLPC still motivate other projects trying to use technology to "disrupt" education and development. Announced in 2005 by MIT Media Lab cofounder Nicholas Negroponte, One Laptop Per Child promised to transform the lives of children across the Global South with a small, sturdy, and cheap laptop computer, powered by a hand crank. In reality, the project fell short in many ways, starting with the hand crank, which never materialized. Yet the project remained charismatic to many who were enchanted by its claims of access to educational opportunities previously out of reach. Behind its promises, OLPC, like many technology projects that make similarly grand claims, had a fundamentally flawed vision of who the computer was made for and what role technology should play in learning.

The ACM constitution provides that our Association holds a general election in the even-numbered years for the positions of President, Vice President, Secretary/Treasurer, and Members-at-Large. Biographical information and statements of the candidates appear on the following pages (candidates' names appear in random order). In addition to the election of ACM's officers--President, Vice President, Secretary/Treasurer--five Members-at-Large will be elected to serve on ACM Council. Please refer to the instructions posted at https://www.esc-vote.com/acm. To access the secure voting site, you will need to enter your email address (the email address associated with your ACM member record) and your unique PIN provided by Election Services Co. Please return your ballot in the enclosed envelope, which must be signed by you on the outside in the space provided. The signed ballot envelope may be inserted into a separate envelope for mailing if you prefer this method. All ballots must be received by no later than 16:00 UTC on 22 May 2020. Validation by the Tellers Committee will take place at 14:00 UTC on 26 May 2020. Elizabeth Churchill is a Director of User Experience at Google. Her field of study is Human Computer Interaction (HCI) and User Experience (UX), with a current focus on the design of effective designer and developer tools. Churchill has built research groups and led research in a number of well-known companies, including as Director of Human Computer Interaction at eBay Research Labs in San Jose, CA, as a Principal Research Scientist and Research Manager at Yahoo! in Santa Clara, CA, and as a Senior Scientist at the Palo Alto Research Center (PARC) and FXPAL, Fuji Xerox's Research lab in Silicon Valley. Working across a number of research areas, she has over 100 peer reviewed top-tier journal and conference publications in theoretical and applied psychology, cognitive science, human-computer interaction, mobile and ubiquitous computing, computer-mediated communication, and social media, more than 50 patents granted or pending, and 7 academic books. Her team produces research that impacts a large number of Google's products (by shaping Google's Flutter and Material Design), influencing the work of hundreds of thousands of designers and developers globally, and thus affecting the user experience of millions of end-users. She continues to guest lecture at universities and to mentor early stage career professionals and students.