By now, everyone's seen the videos of Boston Dynamics robot dog, Spot. It can walk, run, hop on two legs and even dance -- it's mighty impressive. But with every video released by the American robotics firm, it felt like we were edging closer to the ultimate goal of four-legged drones that could be equipped for battle and replace soldiers. However, Boston Dynamics has come together with a coalition of other robotics experts to plead with companies across the sector to please never give the robots guns. The letter, which was first reported by Axios, has been signed by Boston Dynamics, Agility Robotics, ANYbotics, Unitree, Clearpath and Open Robotics.

Companies that use AI in hiring are trying to determine how to comply with a New York law that mandates they test their systems for potential biases. Businesses and their service providers are grappling with how to comply with New York City's mandate for audits of artificial intelligence systems used in hiring. A New York City law that comes into effect in January will require companies to conduct audits to assess biases, including along race and gender lines, in the AI systems they use in hiring. Under New York's law, the hiring company is ultimately liable--and can face fines--for violations. But the requirement has posed some compliance challenges.

Treatments targeting complex diseases, such as cancer, frequently lead to acquired drug resistance, due to patient-specific variability. For instance, drugs targeting only one key component of growth or proliferation pathways, may lead to selective pressure and activation of a compensatory mechanism [1], thus making this treatment suboptimal. However, during multi-target inhibition with reduced stringency, drug resistance is less likely. Therefore, the implementation of combination therapy might improve patient treatment as different drugs may target distinct pathways or genes, likely leading to decreased cancer cell survival. In addition to the increased efficacy, combination therapy often reduces toxicity and decreases the likelihood of treatment resistance compared to monotherapy (i.e., single drug) treatments [2]. Due to advancements in high-throughput screening (HTS), the number of drug screening datasets has been growing in recent years. Some examples include the NCI-ALMANAC dataset [3] which contains 103 FDA-approved drugs tested in 60 different cell lines (NCI-60) [4] or the large oncology dataset produced by Merck&Co [5] which is composed of 38 drugs tested in 39 different cell lines from 6 different tissue types.

Pollutant exposure during gestation is a known and adverse factor for birth and health outcomes. However, the links between prenatal air pollution exposures and educational outcomes are less clear, in particular the critical windows of susceptibility during pregnancy. Using a large cohort of students in North Carolina, we study the link between prenatal daily $\mbox{PM}_{2.5}$ exposure and 4th end-of-grade reading scores. We develop and apply a locally adaptive and highly scalable Bayesian regression model for scalar responses with functional and scalar predictors. The proposed model pairs a B-spline basis expansion with dynamic shrinkage priors to capture both smooth and rapidly-changing features in the regression surface. The model is accompanied by a new decision analysis approach for functional regression that extracts the critical windows of susceptibility and guides the model interpretations. These tools help to identify and address broad limitations with the interpretability of functional regression models. Simulation studies demonstrate more accurate point estimation, more precise uncertainty quantification, and far superior window selection than existing approaches. Leveraging the proposed modeling, computational, and decision analysis framework, we conclude that prenatal $\mbox{PM}_{2.5}$ exposure during early and late pregnancy is most adverse for 4th end-of-grade reading scores.

Code-switching is a common phenomenon among multilingual speakers, where alternation between two or more languages occurs within the context of a single conversation. While multilingual humans can seamlessly switch back and forth between languages, multilingual neural machine translation (NMT) models are not robust to such sudden changes in input. This work explores multilingual NMT models' ability to handle code-switched text. First, we propose checks to measure switching capability. Second, we investigate simple and effective data augmentation methods that can enhance an NMT model's ability to support code-switching. Finally, by using a glass-box analysis of attention modules, we demonstrate the effectiveness of these methods in improving robustness.

While there has been substantial progress in developing systems to automate fact-checking, they still lack credibility in the eyes of the users. Thus, an interesting approach has emerged: to perform automatic fact-checking by verifying whether an input claim has been previously fact-checked by professional fact-checkers and to return back an article that explains their decision. This is a sensible approach as people trust manual fact-checking, and as many claims are repeated multiple times. Yet, a major issue when building such systems is the small number of known tweet--verifying article pairs available for training. Here, we aim to bridge this gap by making use of crowd fact-checking, i.e., mining claims in social media for which users have responded with a link to a fact-checking article. In particular, we mine a large-scale collection of 330,000 tweets paired with a corresponding fact-checking article. We further propose an end-to-end framework to learn from this noisy data based on modified self-adaptive training, in a distant supervision scenario. Our experiments on the CLEF'21 CheckThat! test set show improvements over the state of the art by two points absolute. Our code and datasets are available at https://github.com/mhardalov/crowdchecked-claims

Recent language generative models are mostly trained on large-scale datasets, while in some real scenarios, the training datasets are often expensive to obtain and would be small-scale. In this paper we investigate the challenging task of less-data constrained generation, especially when the generated news headlines are short yet expected by readers to keep readable and informative simultaneously. We highlight the key information modeling task and propose a novel duality fine-tuning method by formally defining the probabilistic duality constraints between key information prediction and headline generation tasks. The proposed method can capture more information from limited data, build connections between separate tasks, and is suitable for less-data constrained generation tasks. Furthermore, the method can leverage various pre-trained generative regimes, e.g., autoregressive and encoder-decoder models. We conduct extensive experiments to demonstrate that our method is effective and efficient to achieve improved performance in terms of language modeling metric and informativeness correctness metric on two public datasets.

Soft prompts have been recently proposed as a tool for adapting large frozen language models (LMs) to new tasks. In this work, we repurpose soft prompts to the task of injecting world knowledge into LMs. We introduce a method to train soft prompts via self-supervised learning on data from knowledge bases. The resulting soft knowledge prompts (KPs) are task independent and work as an external memory of the LMs. We perform qualitative and quantitative experiments and demonstrate that: (1) KPs can effectively model the structure of the training data; (2) KPs can be used to improve the performance of LMs in different knowledge intensive tasks.

Open-source is frequently described as a driver for unprecedented communication and collaboration, and the process works best when projects support teamwork. Yet, open-source cooperation processes in no way protect project contributors from considerations of trust, power, and influence. Indeed, achieving the level of trust necessary to contribute to a project and thus influence its direction is a constant process of change, and developers take many different routes over many communication channels to achieve it. We refer to this process of influence-seeking and trust-building as trust ascendancy. This paper describes a methodology for understanding the notion of trust ascendancy and introduces the capabilities that are needed to localize trust ascendancy operations happening over open-source projects. Much of the prior work in understanding trust in open-source software development has focused on a static view of the problem using different forms of quantity measures. However, trust ascendancy is not static, but rather adapts to changes in the open-source ecosystem in response to new input. This paper is the first attempt to articulate and study these signals from a dynamic view of the problem. In that respect, we identify related work that may help illuminate research challenges, implementation tradeoffs, and complementary solutions. Our preliminary results show the effectiveness of our method at capturing the trust ascendancy developed by individuals involved in a well-documented 2020 social engineering attack. Our future plans highlight research challenges and encourage cross-disciplinary collaboration to create more automated, accurate, and efficient ways to model and then track trust ascendancy in open-source projects.

The ubiquity of Machine Learning (ML) models, and more specifically deep neural network (NN) models, in all sorts of applications has become undeniable in recent years. From classifying images [1, 2, 3], detecting objects [4, 1] and performing semantic segmentation [5, 4] to translating from one human language to another [6] and doing sentiment analysis [7], the advances in different subfields of ML can be attributed mostly to the explosion of computing power and their ability to speed up the training process of artificial NNs. Most famously, AlexNet [8] allowed for an impressive jump in performance in the challenging ILSVRC2012 image classification dataset [1], also known as ImageNet, permanently cementing deep convolutional NN (CNN) architectures in the field of computer vision. Since then, architectures have gotten more refined [9, 10], training procedures have gotten increasingly more complex [11], and their performance and robustness have greatly improved as a consequence. Namely, the success of these deep CNN models is related to their ability to treat high-dimensional and complex data such as images or natural language. The impressive performance of NNs for machine learning tasks can be explained by the ability of their flexible architecture to capture meaningful information on various kinds of complex data and the fact that they are potentially composed of millions of parameters. However, this poses a major challenge: deciphering the reasoning behind the model's predictions. For instance, typical NN architectures for classification or regression problems incrementally transform the representation of the input data in the so-called latent space (or feature space) and then use this transformed representation to make their predictions, as summarized in Figure 1. Each step of this incremental data processing pipeline (or feature extraction chain) is carried out by a so-called layer, which is mathematically a non-linear function (blue rectangle in Figure 1).