Search

We propose misogyny detection as an Argumentative Reasoning task and we investigate the capacity of large language models (LLMs) to understand the implicit reasoning used to convey misogyny in both Italian and English. The central aim is to generate the missing reasoning link between a message and the implied meanings encoding the misogyny. Our study uses argumentation theory as a foundation to form a collection of prompts in both zero-shot and few-shot settings. These prompts integrate different techniques, including chain-of-thought reasoning and augmented knowledge. Our findings show that LLMs fall short on reasoning capabilities about misogynistic comments and that they mostly rely on their implicit knowledge derived from internalized common stereotypes about women to generate implied assumptions, rather than on inductive reasoning.

Robust estimation of the essential matrix, which encodes the relative position and orientation of two cameras, is a fundamental step in structure from motion pipelines. Recent deep-based methods achieved accurate estimation by using complex network architectures that involve graphs, attention layers, and hard pruning steps. Here, we propose a simpler network architecture based on Deep Sets. Given a collection of point matches extracted from two images, our method identifies outlier point matches and models the displacement noise in inlier matches. A weighted DLT module uses these predictions to regress the essential matrix. Our network achieves accurate recovery that is superior to existing networks with significantly more complex architectures.

Multiview Structure from Motion is a fundamental and challenging computer vision problem. A recent deep-based approach was proposed utilizing matrix equivariant architectures for the simultaneous recovery of camera pose and 3D scene structure from large image collections. This work however made the unrealistic assumption that the point tracks given as input are clean of outliers. Here we propose an architecture suited to dealing with outliers by adding an inlier/outlier classifying module that respects the model equivariance and by adding a robust bundle adjustment step. Experiments demonstrate that our method can be successfully applied in realistic settings that include large image collections and point tracks extracted with common heuristics and include many outliers.

Patient-Centric Knowledge Graphs (PCKGs) represent an important shift in healthcare that focuses on individualized patient care by mapping the patient's health information in a holistic and multi-dimensional way. PCKGs integrate various types of health data to provide healthcare professionals with a comprehensive understanding of a patient's health, enabling more personalized and effective care. This literature review explores the methodologies, challenges, and opportunities associated with PCKGs, focusing on their role in integrating disparate healthcare data and enhancing patient care through a unified health perspective. In addition, this review also discusses the complexities of PCKG development, including ontology design, data integration techniques, knowledge extraction, and structured representation of knowledge. It highlights advanced techniques such as reasoning, semantic search, and inference mechanisms essential in constructing and evaluating PCKGs for actionable healthcare insights. We further explore the practical applications of PCKGs in personalized medicine, emphasizing their significance in improving disease prediction and formulating effective treatment plans. Overall, this review provides a foundational perspective on the current state-of-the-art and best practices of PCKGs, guiding future research and applications in this dynamic field.

This paper presents TL;DR Progress, a new tool for exploring the literature on neural text summarization. It organizes 514~papers based on a comprehensive annotation scheme for text summarization approaches and enables fine-grained, faceted search. Each paper was manually annotated to capture aspects such as evaluation metrics, quality dimensions, learning paradigms, challenges addressed, datasets, and document domains. In addition, a succinct indicative summary is provided for each paper, consisting of automatically extracted contextual factors, issues, and proposed solutions. The tool is available online at https://www.tldr-progress.de, a demo video at https://youtu.be/uCVRGFvXUj8, Comment: EACL 2024 System Demonstration

In recent years, the neural radiance field (NeRF) model has gained popularity due to its ability to recover complex 3D scenes. Following its success, many approaches proposed different NeRF representations in order to further improve both runtime and performance. One such example is Triplane, in which NeRF is represented using three 2D feature planes. This enables easily using existing 2D neural networks in this framework, e.g., to generate the three planes. Despite its advantage, the triplane representation lagged behind in its 3D recovery quality compared to NeRF solutions. In this work, we propose TriNeRFLet, a 2D wavelet-based multiscale triplane representation for NeRF, which closes the 3D recovery performance gap and is competitive with current state-of-the-art methods. Building upon the triplane framework, we also propose a novel super-resolution (SR) technique that combines a diffusion model with TriNeRFLet for improving NeRF resolution., Comment: Accepted to ECCV 2024. Webpage link: https://rajaeekh.github.io/trinerflet-web

This comprehensive study delves into the realm of indoor positioning technologies within the domain of Smart Education (SE). Focusing on typical techniques and technologies in educational settings, the research emphasizes the importance and potential services of localization in SE. Moreover, this work explores the feasibility and limitations of these technologies, providing a detailed account of their role in educational settings. The paper also contains in an innovative Proof of Concept (PoC), demonstrating an automatic attendance control (AAC) system that integrates 5G and WiFi technologies. This PoC effectively showcases the possibilities and effectiveness of location-based services in educational surroundings even with a limited budget, setting the stage for optimizing teaching time, enhancing the quality of education., Comment: 14 pages, 11 figures, 2 tables

Network motif identification problem aims to find topological patterns in biological networks. Identifying non-overlapping motifs is a computationally challenging problem using classical computers. Quantum computers enable solving high complexity problems which do not scale using classical computers. In this paper, we develop the first quantum solution, called QOMIC (Quantum Optimization for Motif IdentifiCation), to the motif identification problem. QOMIC transforms the motif identification problem using a integer model, which serves as the foundation to develop our quantum solution. We develop and implement the quantum circuit to find motif locations in the given network using this model. Our experiments demonstrate that QOMIC outperforms the existing solutions developed for the classical computer, in term of motif counts. We also observe that QOMIC can efficiently find motifs in human regulatory networks associated with five neurodegenerative diseases: Alzheimers, Parkinsons, Huntingtons, Amyotrophic Lateral Sclerosis (ALS), and Motor Neurone Disease (MND).

Current approaches to automatic summarization of scientific papers generate informative summaries in the form of abstracts. However, abstracts are not intended to show the relationship between a paper and the references cited in it. We propose a new contextualized summarization approach that can generate an informative summary conditioned on a given sentence containing the citation of a reference (a so-called "citance"). This summary outlines the content of the cited paper relevant to the citation location. Thus, our approach extracts and models the citances of a paper, retrieves relevant passages from cited papers, and generates abstractive summaries tailored to each citance. We evaluate our approach using $\textbf{Webis-Context-SciSumm-2023}$, a new dataset containing 540K~computer science papers and 4.6M~citances therein., Comment: Accepted at EMNLP 2023 Findings

Online forums encourage the exchange and discussion of different stances on many topics. Not only do they provide an opportunity to present one's own arguments, but may also gather a broad cross-section of others' arguments. However, the resulting long discussions are difficult to overview. This paper presents a novel unsupervised approach using large language models (LLMs) to generating indicative summaries for long discussions that basically serve as tables of contents. Our approach first clusters argument sentences, generates cluster labels as abstractive summaries, and classifies the generated cluster labels into argumentation frames resulting in a two-level summary. Based on an extensively optimized prompt engineering approach, we evaluate 19~LLMs for generative cluster labeling and frame classification. To evaluate the usefulness of our indicative summaries, we conduct a purpose-driven user study via a new visual interface called Discussion Explorer: It shows that our proposed indicative summaries serve as a convenient navigation tool to explore long discussions., Comment: Accepted at EMNLP 2023 Main Conference

Existing 3D instance segmentation methods typically assume that all semantic classes to be segmented would be available during training and only seen categories are segmented at inference. We argue that such a closed-world assumption is restrictive and explore for the first time 3D indoor instance segmentation in an open-world setting, where the model is allowed to distinguish a set of known classes as well as identify an unknown object as unknown and then later incrementally learning the semantic category of the unknown when the corresponding category labels are available. To this end, we introduce an open-world 3D indoor instance segmentation method, where an auto-labeling scheme is employed to produce pseudo-labels during training and induce separation to separate known and unknown category labels. We further improve the pseudo-labels quality at inference by adjusting the unknown class probability based on the objectness score distribution. We also introduce carefully curated open-world splits leveraging realistic scenarios based on inherent object distribution, region-based indoor scene exploration and randomness aspect of open-world classes. Extensive experiments reveal the efficacy of the proposed contributions leading to promising open-world 3D instance segmentation performance., Comment: Accepted at NeurIPS 2023

Past work has considered the analytic properties of the reflection coefficient for a metal-backed slab. The primary result established a fundamental relationship for the minimal layer thickness to bandwidth ratio achievable for an absorber. There has yet to be establishment of a similar relationship for non metal-backed layers, and here we present the universal result based on the Kramers-Kronig relations. Our theory is validated with transfer matrix calculations of homogeneous materials, and full-wave numerical simulations of electromagnetic metamaterials. Our results place more general fundamental limits on absorbers and thus will be important for both fundamental and applied studies., Comment: 6 pages, 5 figures (SI not included)

Previous deep learning efforts have focused on improving the performance of Pulmonary Embolism(PE) diagnosis from Computed Tomography (CT) scans using Convolutional Neural Networks (CNN). However, the features from CT scans alone are not always sufficient for the diagnosis of PE. CT scans along with electronic heath records (EHR) can provide a better insight into the patients condition and can lead to more accurate PE diagnosis. In this paper, we propose Pulmonary Embolism Detection using Contrastive Learning (PECon), a supervised contrastive pretraining strategy that employs both the patients CT scans as well as the EHR data, aiming to enhance the alignment of feature representations between the two modalities and leverage information to improve the PE diagnosis. In order to achieve this, we make use of the class labels and pull the sample features of the same class together, while pushing away those of the other class. Results show that the proposed work outperforms the existing techniques and achieves state-of-the-art performance on the RadFusion dataset with an F1-score of 0.913, accuracy of 0.90 and an AUROC of 0.943. Furthermore, we also explore the explainability of our approach in comparison to other methods. Our code is publicly available at https://github.com/BioMedIA-MBZUAI/PECon.

One of the major impediments in deployment of Autonomous Driving Systems (ADS) is their safety and reliability. The primary reason for the complexity of testing ADS is that it operates in an open world characterized by its non-deterministic, high-dimensional and non-stationary nature where the actions of other actors in the environment are uncontrollable from the ADS's perspective. This leads to a state space explosion problem and one way of mitigating this problem is by concretizing the scope for the system under test (SUT) by testing for a set of behavioral competencies which an ADS must demonstrate. A popular approach to testing ADS is scenario-based testing where the ADS is presented with driving scenarios from real world (and synthetically generated) data and expected to meet defined safety criteria while navigating through the scenario. We present SAFR-AV, an end-to-end ADS testing platform to enable scenario-based ADS testing. Our work addresses key real-world challenges of building an efficient large scale data ingestion pipeline and search capability to identify scenarios of interest from real world data, creating digital twins of the real-world scenarios to enable Software-in-the-Loop (SIL) testing in ADS simulators and, identifying key scenario parameter distributions to enable optimization of scenario coverage. These along with other modules of SAFR-AV would allow the platform to provide ADS pre-certifications.

Human pose estimation (HPE) is a classical task in computer vision that focuses on representing the orientation of a person by identifying the positions of their joints. We design a lighterversion of the stacked hourglass network with minimal loss in performance of the model. The lightweight 2-stacked hourglass has a reduced number of channels with depthwise separable convolutions, residual connections with concatenation, and residual connections between the necks of the hourglasses. The final model has a marginal drop in performance with 79% reduction in the number of parameters and a similar drop in MAdds, Comment: Course project

Intra-class variations in the open world lead to various challenges in classification tasks. To overcome these challenges, fine-grained classification was introduced, and many approaches were proposed. Some rely on locating and using distinguishable local parts within images to achieve invariance to viewpoint changes, intra-class differences, and local part deformations. Our approach, which is inspired by P2P-Net, offers an end-to-end trainable attention-based parts alignment module, where we replace the graph-matching component used in it with a self-attention mechanism. The attention module is able to learn the optimal arrangement of parts while attending to each other, before contributing to the global loss., Comment: Course Assignment

Logistics is a key economic sector where any optimization that reduces costs or improves service has a great impact on society at large. In this paper, the role of two 5G Network Slicing (NS) strategies in Smart Logistics is studied: the use of a static slice with a balance division of network resources and the use of a dynamic slice. To validate the potential gains of these strategies, a Distribution Center with 5G connectivity is simulated, recreating the activity that takes place in a real Smart Logistics scenario. Results show that a dynamic slice makes a more efficient usage of the network resources, improving the quality of service for the different traffic profiles, even when there is a traffic peak. This improvement ranges from 6.48\% to 95.65\%, depending on the specific traffic profile and the evaluated metric.

This paper proposes a generalizable, end-to-end deep learning-based method for relative pose regression between two images. Given two images of the same scene captured from different viewpoints, our method predicts the relative rotation and translation (including direction and scale) between the two respective cameras. Inspired by the classical pipeline, our method leverages Image Matching (IM) as a pre-trained task for relative pose regression. Specifically, we use LoFTR, an architecture that utilizes an attention-based network pre-trained on Scannet, to extract semi-dense feature maps, which are then warped and fed into a pose regression network. Notably, we use a loss function that utilizes separate terms to account for the translation direction and scale. We believe such a separation is important because translation direction is determined by point correspondences while the scale is inferred from prior on shape sizes. Our ablations further support this choice. We evaluate our method on several datasets and show that it outperforms previous end-to-end methods. The method also generalizes well to unseen datasets., Comment: Project webpage: https://fadikhatib.github.io/GRelPose

Most research on natural language processing treats bias as an absolute concept: Based on a (probably complex) algorithmic analysis, a sentence, an article, or a text is classified as biased or not. Given the fact that for humans the question of whether a text is biased can be difficult to answer or is answered contradictory, we ask whether an "absolute bias classification" is a promising goal at all. We see the problem not in the complexity of interpreting language phenomena but in the diversity of sociocultural backgrounds of the readers, which cannot be handled uniformly: To decide whether a text has crossed the proverbial line between non-biased and biased is subjective. By asking "Is text X more [less, equally] biased than text Y?" we propose to analyze a simpler problem, which, by its construction, is rather independent of standpoints, views, or sociocultural aspects. In such a model, bias becomes a preference relation that induces a partial ordering from least biased to most biased texts without requiring a decision on where to draw the line. A prerequisite for this kind of bias model is the ability of humans to perceive relative bias differences in the first place. In our research, we selected a specific type of bias in argumentation, the stance bias, and designed a crowdsourcing study showing that differences in stance bias are perceptible when (light) support is provided through training or visual aid., Comment: Accepted at AACL-IJCNLP 2022, Findings Volume

Even the most robust autonomous behaviors can fail. The goal of this research is to both recover and collect data from failures, during autonomous task execution, so they can be prevented in the future. We propose haptic intervention for real-time failure recovery and data collection. Elly is a system that allows for seamless transitions between autonomous robot behaviors and human intervention while collecting sensory information from the human's recovery strategy. The system and our design choices were experimentally validated on a single arm task -- installing a lightbulb in a socket -- and a bimanual task -- screwing a cap on a bottle -- using two 7-DOF manipulators equipped 4-finger grippers. In these examples, Elly achieved over 80% task completion during a total of 40 runs.

Future industrial applications will encompass compelling new use cases requiring stringent performance guarantees over multiple key performance indicators (KPI) such as reliability, dependability, latency, time synchronization, security, etc. Achieving such stringent and diverse service requirements necessitates the design of a special-purpose Industrial Internet of Things (IIoT) network comprising a multitude of specialized functionalities and technological enablers. This article proposes an innovative architecture for such a special-purpose 6G IIoT network incorporating seven functional building blocks categorized into: special-purpose functionalities and enabling technologies. The former consists of Wireless Environment Control, Traffic/Channel Prediction, Proactive Resource Management and End-to-End Optimization functions; whereas the latter includes Synchronization and Coordination, Machine Learning and Artificial Intelligence Algorithms, and Auxiliary Functions. The proposed architecture aims at providing a resource-efficient and holistic solution for the complex and dynamically challenging requirements imposed by future 6G industrial use cases. Selected test scenarios are provided and assessed to illustrate cross-functional collaboration and demonstrate the applicability of the proposed architecture in a wireless IIoT network., Comment: 11 pages, 5 figures

We present a generalized version of the Saturation in the Null Space (SNS) algorithm for the task control of redundant robots when hard inequality constraints are simultaneously present both in the joint and in the Cartesian space. These hard bounds should never be violated, are treated equally and in a unified way by the algorithm, and may also be varied, inserted or deleted online. When a joint/Cartesian bound saturates, the robot redundancy is exploited to continue fulfilling the primary task. If no feasible solution exists, an optimal scaling procedure is applied to enforce directional consistency with the original task. Simulation and experimental results on different robotic systems demonstrate the efficiency of the approach. The proposed algorithm can be viewed as a generic platform that is easily applicable to any robotic application in which robots operate in an unstructured environment and online handling of joint and Cartesian constraints is critical., Comment: 8 pages, 10 figures. This work has been submitted to the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2022 with RA-L option) for possible publication

Motivated by the existence of superconductivity in pyrite-structure CuS$_2$, we explore the possibility of ionic-liquid-gating-induced superconductivity in the proximal antiferromagnetic Mott insulator NiS$_2$. A clear gating-induced transition from a two-dimensional insulating state to a three-dimensional metallic state is observed at positive gate bias on single crystal surfaces. No evidence for superconductivity is observed down to the lowest measured temperature of 0.45 K, however. Based on transport, energy-dispersive X-ray spectroscopy, X-ray photoelectron spectroscopy, atomic force microscopy, and other techniques, we deduce an electrochemical gating mechanism involving a substantial decrease in the S:Ni ratio (over hundreds of nm), which is both non-volatile and irreversible. This is in striking contrast to the reversible, volatile, surface-limited, electrostatic gate effect in pyrite FeS$_2$. We attribute this stark difference in electrochemical vs. electrostatic gating response in NiS$_2$ and FeS$_2$ to the much larger S diffusion coefficient in NiS$_2$, analogous to the different behaviors observed among electrolyte-gated oxides with differing O-vacancy diffusivities. The gating irreversibility, on the other hand, is associated with the lack of atmospheric S; this is in contrast to the better understood oxide case, where electrolysis of atmospheric H$_2$O provides an O reservoir. This study of NiS$_2$ thus provides new insight into electrolyte gating mechanisms in functional materials, in a previously unexplored limit., Comment: 4 figures, included supplement

Deep learning (DL) inverse techniques have increased the speed of artificial electromagnetic material (AEM) design and improved the quality of resulting devices. Many DL inverse techniques have succeeded on a number of AEM design tasks, but to compare, contrast, and evaluate assorted techniques it is critical to clarify the underlying ill-posedness of inverse problems. Here we review state-of-the-art approaches and present a comprehensive survey of deep learning inverse methods and invertible and conditional invertible neural networks to AEM design. We produce easily accessible and rapidly implementable AEM design benchmarks, which offers a methodology to efficiently determine the DL technique best suited to solving different design challenges. Our methodology is guided by constraints on repeated simulation and an easily integrated metric, which we propose expresses the relative ill-posedness of any AEM design problem. We show that as the problem becomes increasingly ill-posed, the neural adjoint with boundary loss (NA) generates better solutions faster, regardless of simulation constraints. On simpler AEM design tasks, direct neural networks (NN) fare better when simulations are limited, while geometries predicted by mixture density networks (MDN) and conditional variational auto-encoders (VAE) can improve with continued sampling and re-simulation.

Numerous physical systems are described by ordinary or partial differential equations whose solutions are given by holomorphic or meromorphic functions in the complex domain. In many cases, only the magnitude of these functions are observed on various points on the purely imaginary jw-axis since coherent measurement of their phases is often expensive. However, it is desirable to retrieve the lost phases from the magnitudes when possible. To this end, we propose a physics-infused deep neural network based on the Blaschke products for phase retrieval. Inspired by the Helson and Sarason Theorem, we recover coefficients of a rational function of Blaschke products using a Blaschke Product Neural Network (BPNN), based upon the magnitude observations as input. The resulting rational function is then used for phase retrieval. We compare the BPNN to conventional deep neural networks (NNs) on several phase retrieval problems, comprising both synthetic and contemporary real-world problems (e.g., metamaterials for which data collection requires substantial expertise and is time consuming). On each phase retrieval problem, we compare against a population of conventional NNs of varying size and hyperparameter settings. Even without any hyper-parameter search, we find that BPNNs consistently outperform the population of optimized NNs in scarce data scenarios, and do so despite being much smaller models. The results can in turn be applied to calculate the refractive index of metamaterials, which is an important problem in emerging areas of material science.

An external magnetic field can be used to remotely control small-scaled robots, making them promising candidates for diverse biomedical and engineering applications. We showed that our magnetically actuated millirobot is highly agile and can perform a variety of locomotive tasks such as pivot walking and tumbling in a horizontal plane. Here, we focus on controlling the locomotion outcomes of this millirobot in the pivot walking mode. A mathematical model of the system is developed and the kinematic model is derived. The role of the sweep and tilt angles in the robot's motion is also investigated. We propose two controllers to regulate the gait of the pivot walker. The first one is a proportional-geometric controller, which determines the correct pivot point that the millirobot should use. Then, it regulates the angular velocity proportionally based on the error between the center of the millirobot and the reference trajectory. The second controller is based on a gradient descent optimization technique, which expresses the control action as an optimization problem. These control algorithms enable the millirobot to generate a stable gait while tracking the desired trajectory. We conduct a set of different experiments and simulation runs to establish the effectiveness of proposed controllers for different sweep and tilt angles in terms of the tracking error. The two controllers exhibit an appropriate performance, but it is observed that gradient descent based controller yields faster convergence time, smaller tracking error, and fewer number of steps. Finally, we perform an extensive experimentally parametric analysis on the effect of the sweep angle, tilt angle, and step time on the tracking error. As we expect, the optimization-based controller outperforms the geometric based controller.

Small-size robots offer access to spaces that are inaccessible to larger ones. This type of access is crucial in applications such as drug delivery, environmental detection, and collection of small samples. However, there are some tasks that are not possible to perform using only one robot including assembly and manufacturing at small scales, manipulation of micro- and nano- objects, and robot-based structuring of small-scale materials. The solution to this problem is to use a group of robots as a system. Thus, we focus on tasks that can be achieved using a group of small-scale robots. These robots are typically externally actuated due to their size limitation. Yet, one faces the challenge of controlling a group of robots using a single global input. We propose a control algorithm to position individual members of a swarm in predefined positions. A single control input applies to the system and moves all robots in the same direction. We also add another control modality by using different length robots. An electromagnetic coil system applied external force and steered the millirobots. This millirobot can move in various modes of motion such as pivot walking and tumbling. We propose two new designs of these millirobots. In the first design, the magnets are placed at the center of body to reduce the magnetic attraction force. In the second design, the millirobots are of identical length with two extra legs acting as the pivot points. This way we vary pivot separation in design to take advantage of variable speed in pivot walking mode while keeping the speed constant in tumbling mode. This paper presents a general algorithm for positional control of n millirobots with different lengths to move them from given initial positions to final desired ones. This method is based on choosing a leader that is fully controllable. Simulations and hardware experiments validate these results.

In this paper, we apply a fusion machine learning method to construct an automatic intrusion detection system. Concretely, we employ the orthogonal variance decomposition technique to identify the relevant features in network traffic data. The selected features are used to build a deep neural network for intrusion detection. The proposed algorithm achieves 100% detection accuracy in identifying DDoS attacks. The test results indicate a great potential of the proposed method., Comment: Accepted at ISNCC 2021

Disordered spring networks can exhibit rigidity transitions, due to either the removal of materials in over-constrained networks or the application of strain in under-constrained ones. While an effective medium theory (EMT) exists for the former, there is none for the latter. We, therefore, formulate an EMT for random regular networks, under-constrained spring networks with purely geometrical disorder, to predict their stiffness via the distribution of tensions. We find a linear dependence of stiffness on strain in the rigid phase and a nontrivial dependence on both the mean and standard deviation of the tension distribution. While EMT does not yield highly accurate predictions of shear modulus due to spatial heterogeneities, the noninvasiveness of this EMT makes it an ideal starting point for experimentalists quantifying the mechanics of such networks., Comment: 10 pages, 8 figures

We present an improved version of the Saturation in the Null Space (SNS) algorithm for redundancy resolution at the velocity level. In addition to hard bounds on joint space motion, we consider also Cartesian box constraints that cannot be violated at any time. The modified algorithm combines all bounds into a single augmented generalised vector and gives equal, highest priority to all inequality constraints. When needed, feasibility of the original task is enforced by the SNS task scaling procedure. Simulation results are reported for a 6R planar robot., Comment: 3 pages, 4 figures, 2021 Italian Conference on Robotics and Intelligent Machines (2021 I-RIM)

Framing a news article means to portray the reported event from a specific perspective, e.g., from an economic or a health perspective. Reframing means to change this perspective. Depending on the audience or the submessage, reframing can become necessary to achieve the desired effect on the readers. Reframing is related to adapting style and sentiment, which can be tackled with neural text generation techniques. However, it is more challenging since changing a frame requires rewriting entire sentences rather than single phrases. In this paper, we study how to computationally reframe sentences in news articles while maintaining their coherence to the context. We treat reframing as a sentence-level fill-in-the-blank task for which we train neural models on an existing media frame corpus. To guide the training, we propose three strategies: framed-language pretraining, named-entity preservation, and adversarial learning. We evaluate respective models automatically and manually for topic consistency, coherence, and successful reframing. Our results indicate that generating properly-framed text works well but with tradeoffs.

This paper introduces Summary Explorer, a new tool to support the manual inspection of text summarization systems by compiling the outputs of 55~state-of-the-art single document summarization approaches on three benchmark datasets, and visually exploring them during a qualitative assessment. The underlying design of the tool considers three well-known summary quality criteria (coverage, faithfulness, and position bias), encapsulated in a guided assessment based on tailored visualizations. The tool complements existing approaches for locally debugging summarization models and improves upon them. The tool is available at https://tldr.webis.de/, Comment: Accepted as system demonstration at EMNLP 2021

This is the second paper devoted to energetic rigidity, in which we apply our formalism to examples in two dimensions: underconstrained random regular spring networks, vertex models, and jammed packings of soft particles. Spring networks and vertex models are both highly underconstrained, and first-order constraint counting does not predict their rigidity, but second-order rigidity does. In contrast, spherical jammed packings are overconstrained and thus first-order rigid, meaning that constraint counting is equivalent to energetic rigidity as long as prestresses in the system are sufficiently small. Aspherical jammed packings on the other hand have been shown to be jammed at hypostaticity, which we use to argue for a modified constraint counting for systems that are energetically rigid at quartic order., Comment: 13 pages, 8 figures. Second of a two-part series

The purpose of an argumentative text is to support a certain conclusion. Yet, they are often omitted, expecting readers to infer them rather. While appropriate when reading an individual text, this rhetorical device limits accessibility when browsing many texts (e.g., on a search engine or on social media). In these scenarios, an explicit conclusion makes for a good candidate summary of an argumentative text. This is especially true if the conclusion is informative, emphasizing specific concepts from the text. With this paper we introduce the task of generating informative conclusions: First, Webis-ConcluGen-21 is compiled, a large-scale corpus of 136,996 samples of argumentative texts and their conclusions. Second, two paradigms for conclusion generation are investigated; one extractive, the other abstractive in nature. The latter exploits argumentative knowledge that augment the data via control codes and finetuning the BART model on several subsets of the corpus. Third, insights are provided into the suitability of our corpus for the task, the differences between the two generation paradigms, the trade-off between informativeness and conciseness, and the impact of encoding argumentative knowledge. The corpus, code, and the trained models are publicly available.

The availability of labelled data is one of the main limitations in machine learning. We can alleviate this using weak supervision: a framework that uses expert-defined rules $\boldsymbol{\lambda}$ to estimate probabilistic labels $p(y|\boldsymbol{\lambda})$ for the entire data set. These rules, however, are dependent on what experts know about the problem, and hence may be inaccurate or may fail to capture important parts of the problem-space. To mitigate this, we propose Active WeaSuL: an approach that incorporates active learning into weak supervision. In Active WeaSuL, experts do not only define rules, but they also iteratively provide the true label for a small set of points where the weak supervision model is most likely to be mistaken, which are then used to better estimate the probabilistic labels. In this way, the weak labels provide a warm start, which active learning then improves upon. We make two contributions: 1) a modification of the weak supervision loss function, such that the expert-labelled data inform and improve the combination of weak labels; and 2) the maxKL divergence sampling strategy, which determines for which data points expert labelling is most beneficial. Our experiments show that when the budget for labelling data is limited (e.g. $\leq 60$ data points), Active WeaSuL outperforms weak supervision, active learning, and competing strategies, with only a handful of labelled data points. This makes Active WeaSuL ideal for situations where obtaining labelled data is difficult., Comment: Accepted to the ICLR 2021 Workshop on Weakly Supervised Learning

We propose an extension of the Cox-Ross-Rubinstein (CRR) model based on $q$-binomial (or Kemp) random walks, with application to default with logistic failure rates. This model allows us to consider time-dependent switching probabilities varying according to a trend parameter on a non-self-similar binomial tree. In particular, it includes tilt and stretch parameters that control increment sizes. Option pricing formulas are written using $q$-binomial coefficients, and we study the convergence of this model to a Black-Scholes type formula in continuous time. A convergence rate of order $O(N^{-1/2})$ is obtained.

We explore the magnetically-ordered ground state of the isovalently-substituted Mott-insulator Y$_{1-x}$La$_{x}$TiO$_{3}$ for $x$ $\leq$ 0.3 via single crystal growth, magnetometry, neutron diffraction, x-ray magnetic circular dichroism (XMCD), muon spin rotation ($\mu$SR) and small-angle neutron scattering (SANS). We find that the decrease in the magnetic transition temperature on approaching the ferromagnetic (FM) - antiferromagnetic (AFM) phase boundary at the La concentration $x_c$ $\approx$ 0.3 is accompanied by a strong suppression of both bulk and local ordered magnetic moments, along with a volume-wise separation into magnetically-ordered and paramagnetic regions. The thermal phase transition does not show conventional second-order behavior, since neither a clear signature of dynamic critical behavior nor a power-law divergence of the magnetic correlation length is found for the studied substitution range; this finding becomes increasingly obvious with substitution. Finally, from SANS and magnetometry measurements, we discern a crossover from easy-axis to easy-plane magneto-crystalline anisotropy with increasing La substitution. These results indicate complex changes in magnetic structure upon approaching the phase boundary., Comment: 16 pages, 16 figures

Rigidity regulates the integrity and function of many physical and biological systems. This is the first of two papers on the origin of rigidity, wherein we propose that "energetic rigidity," in which all non-trivial deformations raise the energy of a structure, is a more useful notion of rigidity in practice than two more commonly used rigidity tests: Maxwell-Calladine constraint counting (first-order rigidity) and second-order rigidity. We find that constraint counting robustly predicts energetic rigidity only when the system has no states of self stress. When the system has states of self stress, we show that second-order rigidity can imply energetic rigidity in systems that are not considered rigid based on constraint counting, and is even more reliable than shear modulus. We also show that there may be systems for which neither first nor second-order rigidity imply energetic rigidity. The formalism of energetic rigidity unifies our understanding of mechanical stability and also suggests new avenues for material design., Comment: 14 pages, 2 figures. First of a two-part series

In this paper we develop the analytical framework for a novel Wireless signal-based Sensing capability for Robotics (WSR) by leveraging robots' mobility. It allows robots to primarily measure relative direction, or Angle-of-Arrival (AOA), to other robots, while operating in non-line-of-sight unmapped environments and without requiring external infrastructure. We do so by capturing all of the paths that a wireless signal traverses as it travels from a transmitting to a receiving robot in the team, which we term as an AOA profile. The key intuition behind our approach is to enable a robot to emulate antenna arrays as it moves freely in 2D and 3D space. The small differences in the phase of the wireless signals are thus processed with knowledge of robots' local displacement to obtain the profile, via a method akin to Synthetic Aperture Radar (SAR). The main contribution of this work is the development of i) a framework to accommodate arbitrary 2D and 3D motion, as well as continuous mobility of both signal transmitting and receiving robots, while computing AOA profiles between them and ii) a Cramer-Rao Bound analysis, based on antenna array theory, that provides a lower bound on the variance in AOA estimation as a function of the geometry of robot motion. We show that allowing robots to use their full mobility in 3D space while performing SAR, results in more accurate AOA profiles and thus better AOA estimation. All analytical developments are substantiated by extensive simulation and hardware experiments on air/ground robot platforms using 5 GHz WiFi. Our experimental results bolster our analytical findings, demonstrating that 3D motion provides enhanced and consistent accuracy, with total AOA error of less than 10 degree for 95% of trials. We also analytically characterize the impact of displacement estimation errors on the measured AOA., Comment: 27 pages, 27 figures, *co-primary authors

Media organizations bear great reponsibility because of their considerable influence on shaping beliefs and positions of our society. Any form of media can contain overly biased content, e.g., by reporting on political events in a selective or incomplete manner. A relevant question hence is whether and how such form of imbalanced news coverage can be exposed. The research presented in this paper addresses not only the automatic detection of bias but goes one step further in that it explores how political bias and unfairness are manifested linguistically. In this regard we utilize a new corpus of 6964 news articles with labels derived from adfontesmedia.com and develop a neural model for bias assessment. By analyzing this model on article excerpts, we find insightful bias patterns at different levels of text granularity, from single words to the whole article discourse.

Media plays an important role in shaping public opinion. Biased media can influence people in undesirable directions and hence should be unmasked as such. We observe that featurebased and neural text classification approaches which rely only on the distribution of low-level lexical information fail to detect media bias. This weakness becomes most noticeable for articles on new events, where words appear in new contexts and hence their "bias predictiveness" is unclear. In this paper, we therefore study how second-order information about biased statements in an article helps to improve detection effectiveness. In particular, we utilize the probability distributions of the frequency, positions, and sequential order of lexical and informational sentence-level bias in a Gaussian Mixture Model. On an existing media bias dataset, we find that the frequency and positions of biased statements strongly impact article-level bias, whereas their exact sequential order is secondary. Using a standard model for sentence-level bias detection, we provide empirical evidence that article-level bias detectors that use second-order information clearly outperform those without.

The fields of signal and image processing have been deeply influenced by the introduction of deep neural networks. These are successfully deployed in a wide range of real-world applications, obtaining state of the art results and surpassing well-known and well-established classical methods. Despite their impressive success, the architectures used in many of these neural networks come with no clear justification. As such, these are usually treated as "black box" machines that lack any kind of interpretability. A constructive remedy to this drawback is a systematic design of such networks by unfolding well-understood iterative algorithms. A popular representative of this approach is the Iterative Shrinkage-Thresholding Algorithm (ISTA) and its learned version -- LISTA, aiming for the sparse representations of the processed signals. In this paper we revisit this sparse coding task and propose an unfolded version of a greedy pursuit algorithm for the same goal. More specifically, we concentrate on the well-known Orthogonal-Matching-Pursuit (OMP) algorithm, and introduce its unfolded and learned version. Key features of our Learned Greedy Method (LGM) are the ability to accommodate a dynamic number of unfolded layers, and a stopping mechanism based on representation error, both adapted to the input. We develop several variants of the proposed LGM architecture and test some of them in various experiments, demonstrating their flexibility and efficiency.

We present C and X-band radio observations of the famous utraluminous X-ray source (ULX) Holmberg IX X-1, previously discovered to be associated with an optical emission line nebula several hundred pc in extent. Our recent infrared study of the ULX suggested that a jet could be responsible for the infrared excess detected at the ULX position. The new radio observations, performed using the Karl G. Jansky Very Large Array (VLA) in B-configuration, reveal the presence of a radio counterpart to the nebula with a spectral slope of -0.56 similar to other ULXs. Importantly, we find no evidence for an unresolved radio source associated with the ULX itself, and we set an upper limit on any 5 GHz radio core emission of 6.6 $\mu$Jy ($4.1\times10^{32}$ erg s$^{-1}$). This is 20 times fainter than what we expect if the bubble is energized by a jet. If a jet exists its core component is unlikely to be responsible for the infrared excess unless it is variable. Strong winds which are expected in super-Eddington sources could also play an important role in inflating the radio bubble. We discuss possible interpretations of the radio/optical bubble and we prefer the jet+winds-blown bubble scenario similar to the microquasar SS 433., Comment: Accepted by ApJ. This paper contains the first CATTERPLOT ever published in ApJ!

ML4Chem is an open-source machine learning library for chemistry and materials science. It provides an extendable platform to develop and deploy machine learning models and pipelines and is targeted to the non-expert and expert users. ML4Chem follows user-experience design and offers the needed tools to go from data preparation to inference. Here we introduce its atomistic module for the implementation, deployment, and reproducibility of atom-centered models. This module is composed of six core building blocks: data, featurization, models, model optimization, inference, and visualization. We present their functionality and easiness of use with demonstrations utilizing neural networks and kernel ridge regression algorithms., Comment: 32 pages, 11 Figures

In this paper, we propose a new Robust Nonlinear Quadratic Gaussian (RNQG) controller based on State-Dependent Riccati Equation (SDRE) scheme for continuous-time nonlinear systems. Existing controllers do not account for combined noise and disturbance acting on the system. The proposed controller is based on a Lyapunov function and a cost function includes states, inputs, outputs, disturbance, and the noise acting on the system. We express the RNQG control law in the form of a traditional Riccati equation. Real-time applications of the controller place high computational burden on system implementation. This is mainly due to the nonlinear and complex form of the cost function. In order to solve this problem, this cost function is approximated by a weighted polynomial. The weights are found by using a least-squares technique and a neural network. The approximate cost function is incorporated into the controller by employing a method based on Bellman's principle of optimality. Finally, an inertially stabilized inverted pendulum example is used to verify the utility of the proposed control approach., Comment: 10 pages, 4 Figures, sumbitted to Automatica

Ferroic materials are well known to exhibit heterogeneity in the form of domain walls. Understanding the properties of these boundaries is crucial for controlling functionality with external stimuli and for realizing their potential for ultra-low power memory and logic devices as well as novel computing architectures. In this work, we employ synchrotron-based near-field infrared nano-spectroscopy to reveal the vibrational properties of ferroelastic (90$^\circ$ ferroelectric) domain walls in the hybrid improper ferroelectric Ca$_3$Ti$_2$O$_7$. By locally mapping the Ti-O stretching and Ti-O-Ti bending modes, we reveal how structural order parameters rotate across a wall. Thus, we link observed near-field amplitude changes to underlying structural modulations and test ferroelectric switching models against real space measurements of local structure. This initiative opens the door to broadband infrared nano-imaging of heterogeneity in ferroics.

To achieve a successful grasp, gripper attributes such as its geometry and kinematics play a role as important as the object geometry. The majority of previous work has focused on developing grasp methods that generalize over novel object geometry but are specific to a certain robot hand. We propose UniGrasp, an efficient data-driven grasp synthesis method that considers both the object geometry and gripper attributes as inputs. UniGrasp is based on a novel deep neural network architecture that selects sets of contact points from the input point cloud of the object. The proposed model is trained on a large dataset to produce contact points that are in force closure and reachable by the robot hand. By using contact points as output, we can transfer between a diverse set of multifingered robotic hands. Our model produces over 90% valid contact points in Top10 predictions in simulation and more than 90% successful grasps in real world experiments for various known two-fingered and three-fingered grippers. Our model also achieves 93%, 83% and 90% successful grasps in real world experiments for an unseen two-fingered gripper and two unseen multi-fingered anthropomorphic robotic hands., Comment: Accepted to IEEE Robotics and Automation Letters with ICRA 2020 option

This paper presents a framework which enables a user to more easily make corrections to adversarial texts. While attack algorithms have been demonstrated to automatically build adversaries, changes made by the algorithms can often have poor semantics or syntax. Our framework is designed to facilitate human intervention by aiding users in making corrections. The framework extends existing attack algorithms to work within an evolutionary attack process paired with a visual analytics loop. Using an interactive dashboard a user is able to review the generation process in real time and receive suggestions from the system for edits to be made. The adversaries can be used to both diagnose robustness issues within a single classifier or to compare various classifier options. With the weaknesses identified, the framework can also be used as a first step in mitigating adversarial threats. The framework can be used as part of further research into defense methods in which the adversarial examples are used to evaluate new countermeasures. We demonstrate the framework with a word swapping attack for the task of sentiment classification.