Contextualizing language technologies beyond a single language kindled embracing multiple modalities and languages. Individually, each of these directions undoubtedly proliferated into several NLP tasks. Despite this momentum, most of the multimodal research is primarily centered around English and multilingual research is primarily centered around contexts from text modality. Challenging this conventional setup, researchers studied the unification of multilingual and multimodal (MultiX) streams. The main goal of this work is to catalogue and characterize these works by charting out the categories of tasks, datasets and methods to address MultiX scenarios. To this end, we review the languages studied, gold or silver data with parallel annotations, and understand how these modalities and languages interact in modeling. We present an account of the modeling approaches along with their strengths and weaknesses to better understand what scenarios they can be used reliably. Following this, we present the high-level trends in the overall paradigm of the field. Finally, we conclude by presenting a road map of challenges and promising research directions.

Multi-domain data is becoming increasingly common and presents both challenges and opportunities in the data science community. The integration of distinct data-views can be used for exploratory data analysis, and benefit downstream analysis including machine learning related tasks. With this in mind, we present a novel manifold alignment method called MALI (Manifold alignment with label information) that learns a correspondence between two distinct domains. MALI can be considered as belonging to a middle ground between the more commonly addressed semi-supervised manifold alignment problem with some known correspondences between the two domains, and the purely unsupervised case, where no known correspondences are provided. To do this, MALI learns the manifold structure in both domains via a diffusion process and then leverages discrete class labels to guide the alignment. By aligning two distinct domains, MALI recovers a pairing and a common representation that reveals related samples in both domains. Additionally, MALI can be used for the transfer learning problem known as domain adaptation. We show that MALI outperforms the current state-of-the-art manifold alignment methods across multiple datasets.

We propose the first character-level white-box adversarial attack method against transformer models. The intuition of our method comes from the observation that words are split into subtokens before being fed into the transformer models and the substitution between two close subtokens has a similar effect to the character modification. Our method mainly contains three steps. First, a gradient-based method is adopted to find the most vulnerable words in the sentence. Then we split the selected words into subtokens to replace the origin tokenization result from the transformer tokenizer. Finally, we utilize an adversarial loss to guide the substitution of attachable subtokens in which the Gumbel-softmax trick is introduced to ensure gradient propagation. Meanwhile, we introduce the visual and length constraint in the optimization process to achieve minimum character modifications. Extensive experiments on both sentence-level and token-level tasks demonstrate that our method could outperform the previous attack methods in terms of success rate and edit distance. Furthermore, human evaluation verifies our adversarial examples could preserve their origin labels.

Explicit decomposition modeling, which involves breaking down complex tasks into more straightforward and often more interpretable sub-tasks, has long been a central theme in developing robust and interpretable NLU systems. However, despite the many datasets and resources built as part of this effort, the majority have small-scale annotations and limited scope, which is insufficient to solve general decomposition tasks. In this paper, we look at large-scale intermediate pre-training of decomposition-based transformers using distant supervision from comparable texts, particularly large-scale parallel news. We show that with such intermediate pre-training, developing robust decomposition-based models for a diverse range of tasks becomes more feasible. For example, on semantic parsing, our model, DecompT5, improves 20% to 30% on two datasets, Overnight and TORQUE, over the baseline language model. We further use DecompT5 to build a novel decomposition-based QA system named DecompEntail, improving over state-of-the-art models, including GPT-3, on both HotpotQA and StrategyQA by 8% and 4%, respectively.

Russia has suspended its participation in a landmark agreement that allowed vital grain exports from Ukraine after what it said was a drone attack on Russian ships in occupied Crimea. Russia's defence ministry said Ukraine attacked the Black Sea Fleet near Sevastopol in the annexed Crimean Peninsula with 16 drones in the early hours of Saturday, and that British navy "specialists" had helped coordinate the "terrorist" attack. London bluntly rejected Moscow's claim. The Turkey and UN-brokered deal to unlock grain exports signed between Russia and Ukraine in July is critical to easing the global food crisis caused by the conflict. The agreement has already allowed more than 9 million tonnes of Ukrainian grain to be exported and was due to be renewed on November 19.

Sign up to receive the Future Tense newsletter every other Saturday. The U.N.'s COP27 climate summit kicks off on Nov. 6 in Egypt, inviting us, once again, to consider whether we're doing enough, fast enough, to stave off climate chaos and the suffering that will come with it. The scale of change required is head-spinningly drastic, so even unexpectedly rapid expansions in clean energy won't do much to curb malaise and doomsaying. Here in the U.S., the Inflation Reduction Act, the biggest climate investment in the nation's history, has been met, largely, with collective indifference, despite positive buzz about its potential effectiveness. The bill was, predictably, passed without any Republican votes, a grim reminder of the scale of climate denialism.

Universities around the world are taking steps alongside major technology companies to explore ways to bolster ethics education in the artificial intelligence field in line with an initiative supported by the Vatican. The effort seeks to help those already working or aspiring to work in the tech fields understand that the development of artificial intelligence, or AI, should benefit humanity rather than pose uncontrollable challenges to human life. Participants at a global summit at the University of Notre Dame Oct. 25-26 explored ways to encompass ethics education in coursework with speakers calling for widespread integration in both technical and nontechnical curricula. Casey Fiesler, associate professor of information science at the University of Colorado, told in person and online attendees in a session that the long-held view that ethical topics are a "specialization" within technology education must be put aside. "We should not be teaching ethics in the context of computing so that it is completely separate from everything else that we are doing," Fiesler said in calling for a culture shift in higher education that can reach across society.

Federated learning (FL) aims at optimizing a shared global model over multiple edge devices without transmitting (private) data to the central server. While it is theoretically well-known that FL yields an optimal model -- centrally trained model assuming availability of all the edge device data at the central server -- under mild condition, in practice, it often requires massive amount of iterations until convergence, especially under presence of statistical/computational heterogeneity. This paper utilizes cyclic learning rate at the server side to reduce the number of training iterations with increased performance without any additional computational costs for both the server and the edge devices. Numerical results validate that, simply plugging-in the proposed cyclic aggregation to the existing FL algorithms effectively reduces the number of training iterations with improved performance.

We introduce a new Slovak masked language model called SlovakBERT. This is to our best knowledge the first paper discussing Slovak transformers-based language models. We evaluate our model on several NLP tasks and achieve state-of-the-art results. This evaluation is likewise the first attempt to establish a benchmark for Slovak language models. We publish the masked language model, as well as the fine-tuned models for part-of-speech tagging, sentiment analysis and semantic textual similarity.

Uncertainty on human behaviors poses a significant challenge to autonomous driving in crowded urban environments. The partially observable Markov decision processes (POMDPs) offer a principled framework for planning under uncertainty, often leveraging Monte Carlo sampling to achieve online performance for complex tasks. However, sampling also raises safety concerns by potentially missing critical events. To address this, we propose a new algorithm, LEarning Attention over Driving bEhavioRs (LEADER), that learns to attend to critical human behaviors during planning. LEADER learns a neural network generator to provide attention over human behaviors in real-time situations. It integrates the attention into a belief-space planner, using importance sampling to bias reasoning towards critical events. To train the algorithm, we let the attention generator and the planner form a min-max game. By solving the min-max game, LEADER learns to perform risk-aware planning without human labeling.