Search

Quality control of structures segmentation in volumetric medical images is important for identifying segmentation errors in clinical practice and for facilitating model development. This paper introduces SegQC, a novel framework for segmentation quality estimation and segmentation error detection. SegQC computes an estimate measure of the quality of a segmentation in volumetric scans and in their individual slices and identifies possible segmentation error regions within a slice. The key components include: 1. SegQC-Net, a deep network that inputs a scan and its segmentation mask and outputs segmentation error probabilities for each voxel in the scan; 2. three new segmentation quality metrics, two overlap metrics and a structure size metric, computed from the segmentation error probabilities; 3. a new method for detecting possible segmentation errors in scan slices computed from the segmentation error probabilities. We introduce a new evaluation scheme to measure segmentation error discrepancies based on an expert radiologist corrections of automatically produced segmentations that yields smaller observer variability and is closer to actual segmentation errors. We demonstrate SegQC on three fetal structures in 198 fetal MRI scans: fetal brain, fetal body and the placenta. To assess the benefits of SegQC, we compare it to the unsupervised Test Time Augmentation (TTA)-based quality estimation. Our studies indicate that SegQC outperforms TTA-based quality estimation in terms of Pearson correlation and MAE for fetal body and fetal brain structures segmentation. Our segmentation error detection method achieved recall and precision rates of 0.77 and 0.48 for fetal body, and 0.74 and 0.55 for fetal brain segmentation error detection respectively. SegQC enhances segmentation metrics estimation for whole scans and individual slices, as well as provides error regions detection., Comment: 28 pages, 9 figures

Lexical-semantic resources (LSRs), such as online lexicons or wordnets, are fundamental for natural language processing applications. In many languages, however, such resources suffer from quality issues: incorrect entries, incompleteness, but also, the rarely addressed issue of bias towards the English language and Anglo-Saxon culture. Such bias manifests itself in the absence of concepts specific to the language or culture at hand, the presence of foreign (Anglo-Saxon) concepts, as well as in the lack of an explicit indication of untranslatability, also known as cross-lingual \emph{lexical gaps}, when a term has no equivalent in another language. This paper proposes a novel crowdsourcing methodology for reducing bias in LSRs. Crowd workers compare lexemes from two languages, focusing on domains rich in lexical diversity, such as kinship or food. Our LingoGap crowdsourcing tool facilitates comparisons through microtasks identifying equivalent terms, language-specific terms, and lexical gaps across languages. We validated our method by applying it to two case studies focused on food-related terminology: (1) English and Arabic, and (2) Standard Indonesian and Banjarese. These experiments identified 2,140 lexical gaps in the first case study and 951 in the second. The success of these experiments confirmed the usability of our method and tool for future large-scale lexicon enrichment tasks.

Motivated by the recent appearance of the trillium lattice in the search for materials hosting spin liquids, we study the ground state of the classical Heisenberg model on its linegraph, the trilline lattice. We find that this network realises the recently proposed notion of a fragile spin liquid in three dimensions. Additionally, we analyze the Ising case and argue for a possible $\mathbb{Z}_2$ quantum spin liquid phase in the corresponding quantum dimer model. Like the well-known $U(1)$ spin liquids, the classical phase hosts moment fractionalisation evidenced in the diluted lattice, but unlike these, it exhibits exponential decay both in spin correlations and interactions between fractionalised moments. This provides the first instance of a purely short-range correlated classical Heisenberg spin liquid in three dimensions.

Data and derived information about target victims has always been key for successful attacks, both during historical wars and modern cyber wars. Ours turns out to be an era in which modern cars generate a plethora of data about their drivers, and such data could be extremely attractive for offenders. This paper seeks to assess how well modern cars protect their drivers' data. It pursues its goal at a requirement level by analysing the gaps of the privacy policies of chief automakers such as BMW and Mercedes with respect to the General Data Protection Regulation (GDPR). It is found that both brands are still imprecise about how they comply with a number of GDPR articles, hence compliance often results non-verifiable. Most importantly, while BMW exhibits slightly broader compliance, both brands still fail to comply with a number of relevant articles of the regulation. An interpretation of these findings is a non-negligible likelihood that your car may turn against you should cyberwarfare break out.

The first very long baseline interferometry (VLBI) detections at 870$\mu$m wavelength (345$\,$GHz frequency) are reported, achieving the highest diffraction-limited angular resolution yet obtained from the surface of the Earth, and the highest-frequency example of the VLBI technique to date. These include strong detections for multiple sources observed on inter-continental baselines between telescopes in Chile, Hawaii, and Spain, obtained during observations in October 2018. The longest-baseline detections approach 11$\,$G$\lambda$ corresponding to an angular resolution, or fringe spacing, of 19$\mu$as. The Allan deviation of the visibility phase at 870$\mu$m is comparable to that at 1.3$\,$mm on the relevant integration time scales between 2 and 100$\,$s. The detections confirm that the sensitivity and signal chain stability of stations in the Event Horizon Telescope (EHT) array are suitable for VLBI observations at 870$\mu$m. Operation at this short wavelength, combined with anticipated enhancements of the EHT, will lead to a unique high angular resolution instrument for black hole studies, capable of resolving the event horizons of supermassive black holes in both space and time., Comment: Corresponding author: S. Doeleman

In this paper we propose novel global and regional stability analysis conditions based on linear matrix inequalities for a general class of recurrent neural networks. These conditions can be also used for state-feedback control design and a suitable optimization problem enforcing H2 norm minimization properties is defined. The theoretical results are corroborated by numerical simulations, showing the advantages and limitations of the methods presented herein.

We determine necessary and sufficient conditions for unicritical polynomials to be dynamically irreducible over finite fields. This result extends the results of Boston-Jones and Hamblen-Jones-Madhu regarding the dynamical irreducibility of particular families of unicritical polynomials. We also investigate dynamical irreducibility conditions for cubic and shifted linearized polynomials.

This study explores compounding impacts of climate change on power system's load and generation, emphasising the need to integrate adaptation and mitigation strategies into investment planning. We combine existing and novel empirical evidence to model impacts on: i) air-conditioning demand; ii) thermal power outages; iii) hydro-power generation shortages. Using a power dispatch and capacity expansion model, we analyse the Italian power system's response to these climate impacts in 2030, integrating mitigation targets and optimising for cost-efficiency at an hourly resolution. We outline different meteorological scenarios to explore the impacts of both average climatic changes and the intensification of extreme weather events. We find that addressing extreme weather in power system planning will require an extra 5-8 GW of photovoltaic (PV) capacity, on top of the 50 GW of the additional solar PV capacity required by the mitigation target alone. Despite the higher initial investments, we find that the adoption of renewable technologies, especially PV, alleviates the power system's vulnerability to climate change and extreme weather events. Furthermore, enhancing short-term storage with lithium-ion batteries is crucial to counterbalance the reduced availability of dispatchable hydro generation.

Recently, the authors with Lea Beneish established a recipe for constructing Ramanujan-Sato series for $1/\pi$, and used this to construct 11 explicit examples of Ramanujan-Sato series arising from modular forms for arithmetic triangle groups of non-compact type. Here, we use work of Chisholm, Deines, Long, Nebe and the third author to prove a general $p$-adic supercongruence theorem through an explicit connection to CM hypergeometric elliptic curves that provides $p$-adic analogues of these Ramanujan-Sato series. We further use this theorem to construct explicit examples related to each of our explicit Ramanujan-Sato series examples.

Polynomial neural networks have been implemented in a range of applications and present an advantageous framework for theoretical machine learning. A polynomial neural network of fixed architecture and activation degree gives an algebraic map from the network's weights to a set of polynomials. The image of this map is the space of functions representable by the network. Its Zariski closure is an affine variety known as a neurovariety. The dimension of a polynomial neural network's neurovariety provides a measure of its expressivity. In this work, we introduce the notion of the activation threshold of a network architecture which expresses when the dimension of a neurovariety achieves its theoretical maximum. In addition, we prove expressiveness results for polynomial neural networks with equi-width~architectures., Comment: 13 pages

Since the introduction of the self-attention mechanism and the adoption of the Transformer architecture for Computer Vision tasks, the Vision Transformer-based architectures gained a lot of popularity in the field, being used for tasks such as image classification, object detection and image segmentation. However, efficiently leveraging the attention mechanism in vision transformers for the Monocular 3D Object Detection task remains an open question. In this paper, we present LAM3D, a framework that Leverages self-Attention mechanism for Monocular 3D object Detection. To do so, the proposed method is built upon a Pyramid Vision Transformer v2 (PVTv2) as feature extraction backbone and 2D/3D detection machinery. We evaluate the proposed method on the KITTI 3D Object Detection Benchmark, proving the applicability of the proposed solution in the autonomous driving domain and outperforming reference methods. Moreover, due to the usage of self-attention, LAM3D is able to systematically outperform the equivalent architecture that does not employ self-attention., Comment: 6 pages. Accepted to MMSP 2024

Despite their apparent simplicity, devices like smart light bulbs and electrical plugs are often perceived as exempt from rigorous security measures. However, this paper challenges this misconception, uncovering how vulnerabilities in these seemingly innocuous devices can expose users to significant risks. This paper extends the findings outlined in previous work, introducing a novel attack scenario. This new attack allows malicious actors to obtain sensitive credentials, including the victim's Tapo account email and password, as well as the SSID and password of her local network. Furthermore, we demonstrate how these findings can be replicated, either partially or fully, across other smart devices within the same IoT ecosystem, specifically those manufactured by Tp-Link. Our investigation focused on the Tp-Link Tapo range, encompassing smart bulbs (Tapo L530E, Tapo L510E V2, and Tapo L630), a smart plug (Tapo P100), and a smart camera (Tapo C200). Utilizing similar communication protocols, or slight variants thereof, we found that the Tapo L530E, Tapo L510E V2, and Tapo L630 are susceptible to complete exploitation of all attack scenarios, including the newly identified one. Conversely, the Tapo P100 and Tapo C200 exhibit vulnerabilities to only a subset of attack scenarios. In conclusion, by highlighting these vulnerabilities and their potential impact, we aim to raise awareness and encourage proactive steps towards mitigating security risks in smart device deployment.

The theory of nilpotent orbits of simple Lie algebras has seen tremendous developments over the past decades. In this context an important role is played by the component group of the stabilizer of a nilpotent element. In this work, the aim is to show computational methods to obtain explicit generators of the component group of the centralizer of a nilpotent element in a simple Lie algebra over $\mathbb{C}$. In some cases such generators had already been determined by impressive hand calculations but these often use the specific form of the chosen representative and are thus not immediately applicable to different representatives of the same orbit. It is then interesting to show how to overcome this issue constructing specific algorithms: for the classical types there is a straightforward method that directly translates well-known theoretical constructions; for the exceptional types we devise a method using the double centralizer of an $\mathfrak{sl}_2$-triple. In particular, it gives an independent construction of the component group, that does not depend on the prior knowledge of its isomorphism type. For our purposes, we needed to construct many algorithms which have been mainly implemented in the computational algebra system GAP.

The current standard model of cosmology successfully describes a variety of measurements, but the nature of its main ingredients, dark matter and dark energy, remains unknown. Euclid is a medium-class mission in the Cosmic Vision 2015-2025 programme of the European Space Agency (ESA) that will provide high-resolution optical imaging, as well as near-infrared imaging and spectroscopy, over about 14,000 deg^2 of extragalactic sky. In addition to accurate weak lensing and clustering measurements that probe structure formation over half of the age of the Universe, its primary probes for cosmology, these exquisite data will enable a wide range of science. This paper provides a high-level overview of the mission, summarising the survey characteristics, the various data-processing steps, and data products. We also highlight the main science objectives and expected performance., Comment: Accepted for publication in the A&A special issue`Euclid on Sky'

Language technology has the potential to facilitate intercultural communication through meaningful translations. However, the current state of language technology is deeply entangled with colonial knowledge due to path dependencies and neo-colonial tendencies in the global governance of artificial intelligence (AI). Language technology is a complex and emerging field that presents challenges for co-design interventions due to enfolding in assemblages of global scale and diverse sites and its knowledge intensity. This paper uses LiveLanguage, a lexical database, a set of services with particular emphasis on modelling language diversity and integrating small and minority languages, as an example to discuss and close the gap from pluriversal design theory to practice. By diversifying the concept of emerging technology, we can better approach language technology in global contexts. The paper presents a model comprising of five layers of technological activity. Each layer consists of specific practices and stakeholders, thus provides distinctive spaces for co-design interventions as mode of inquiry for de-linking, re-thinking and re-building language technology towards pluriversality. In that way, the paper contributes to reflecting the position of co-design in decolonising emergent technologies, and to integrating complex theoretical knowledge towards decoloniality into language technology design.

This paper presents DeepKalPose, a novel approach for enhancing temporal consistency in monocular vehicle pose estimation applied on video through a deep-learning-based Kalman Filter. By integrating a Bi-directional Kalman filter strategy utilizing forward and backward time-series processing, combined with a learnable motion model to represent complex motion patterns, our method significantly improves pose accuracy and robustness across various conditions, particularly for occluded or distant vehicles. Experimental validation on the KITTI dataset confirms that DeepKalPose outperforms existing methods in both pose accuracy and temporal consistency., Comment: 4 pages, 3 Figures, published to IET Electronic Letters

The popular subword tokenizers of current language models, such as Byte-Pair Encoding (BPE), are known not to respect morpheme boundaries, which affects the downstream performance of the models. While many improved tokenization algorithms have been proposed, their evaluation and cross-comparison is still an open problem. As a solution, we propose a combined intrinsic-extrinsic evaluation framework for subword tokenization. Intrinsic evaluation is based on our new UniMorph Labeller tool that classifies subword tokenization as either morphological or alien. Extrinsic evaluation, in turn, is performed via the Out-of-Vocabulary Generalization Challenge 1.0 benchmark, which consists of three newly specified downstream text classification tasks. Our empirical findings show that the accuracy of UniMorph Labeller is 98%, and that, in all language models studied (including ALBERT, BERT, RoBERTa, and DeBERTa), alien tokenization leads to poorer generalizations compared to morphological tokenization for semantic compositionality of word meanings.

Excess mortality is defined as an increase in the number of deaths above what is expected based on historical trends, hereafter called baseline. In a previous paper, we introduced a statistical method that allows an unbiased and robust determination of the baseline to be used for the computation of excesses. A good determination of the baseline allows us to efficiently evaluate the excess of casualties that occurred in Italy in the last 12 years and in particular in the last 3 years due to the Coronavirus Disease 2019 (COVID-19) epidemic. To this extent, we have analyzed the data on mortality in Italy in the period January 1st 2011 to December 31th 2022, provided by the Italian National Institute of Statistics (ISTAT). The dataset contains information on deaths for all possible causes, without specific reference to any particular one. The data exhibit strikingly evident periodicity in the number of deaths with pronounced maxima in the winter and minima in the summer, repeating itself in amplitude along the whole twelve-year sample. Superimposed on this wave-like structure are often present excesses of casualties, most likely due to occasional causes of death such as the flu epidemics (in winter) and heat waves (in summer). The very accurate periodicity along the seasons (the "baseline"), allows us to determine with great accuracy and confidence the number of expected deaths for each day of the year in the absence of occasional contributions. Each of the latter can be modeled with an additional function that parameterizes the deviation from the baseline.

We define a topological space to be an "SDL space" if the closure of each one of its strongly discrete subsets is Lindel\"of. After distinguishing this property from the Lindel\"of property we make various remarks about cardinal invariants of SDL spaces. For example we prove that $|X| \leq 2^{\chi(X)}$ for every SDL Urysohn space and that every SDL $P$-space of character $\leq \omega_1$ is regular and has cardinality $\leq 2^{\omega_1}$. Finally, we exploit our results to obtain some partial answers to questions about the cardinality of cellular-Lindel\"of spaces., Comment: Dedicated to the memory of Phil Zenor

High-quality WordNets are crucial for achieving high-quality results in NLP applications that rely on such resources. However, the wordnets of most languages suffer from serious issues of correctness and completeness with respect to the words and word meanings they define, such as incorrect lemmas, missing glosses and example sentences, or an inadequate, Western-centric representation of the morphology and the semantics of the language. Previous efforts have largely focused on increasing lexical coverage while ignoring other qualitative aspects. In this paper, we focus on the Arabic language and introduce a major revision of the Arabic WordNet that addresses multiple dimensions of lexico-semantic resource quality. As a result, we updated more than 58% of the synsets of the existing Arabic WordNet by adding missing information and correcting errors. In order to address issues of language diversity and untranslatability, we also extended the wordnet structure by new elements: phrasets and lexical gaps.

During the querelle des infiniment petits, Leibniz wrote several texts to justify using Differential calculus among Parisian savants. However, only three were published. Among these publications, ''Sentiment de Monsieur Leibnitz'' had a peculiar destiny. Although we are aware of the manuscript (Gotha FB A 448--449, Bl. 41--42), it is only recently that we have identified a copy of its impression in the British Library catalogue. This copy was printed in 1706 together with writings by other mathematicians united in the defence of the new calculus -- Joseph Saurin, Jacob Hermann and the Bernoulli brothers. Recently published epistolary exchanges indicate that Jean-Paul Bignon, at the time director of the Royal Academy of Sciences, in order to calm down the institution, had prohibited this publication and confiscated the prints.This article examine the epistemological and institutional issues at stake in ''Sentiment de Monsieur Leibnitz''.

Criminal investigations are inherently complex as they typically involve interactions among various actors like investigators, prosecutors, and defendants. The pervasive integration of technology in daily life adds an extra layer of complexity, especially in crimes that involve a digital element. The establishment of digital forensics as a foundational discipline for extracting digital evidence further exacerbates the complex nature of criminal investigations, leading to the proliferation of multiple scenarios. Recognising the need to structure standard operating procedures for the handling of digital evidence, the representation of digital forensics as a protocol emerges as a valuable opportunity to identify security and privacy threats. In this paper, we delineate the protocols that compose digital forensics within a criminal case, formalise them as message sequence charts (MSCs), and identify their functional requirements., Comment: Accepted in 12th International Symposium on Digital Forensics and Security (ISDFS 2024). 979-8-3503-3036-6/24/$31.00 copyright 2024 IEEE

We prove quantitative decay estimates for the boundary layer corrector in stochastic homogenization in the case of a half-space boundary. Our estimates are of optimal order and show that the gradient of the boundary layer corrector features nearly fluctuation-order decay; its expected value decays even one order faster. As a corollary, we deduce estimates on the accuracy of the representative volume element method for the computation of effective coefficients: our understanding of the decay of boundary layers enables us to improve the order of convergence of the RVE method for $d\geq 3$.