Search

Software testing is an essential part of the software development cycle to improve the code quality. Typically, a unit test consists of a test prefix and a test oracle which captures the developer's intended behaviour. A known limitation of traditional test generation techniques (e.g. Randoop and Evosuite) is that they produce test oracles that capture the actual program behaviour rather than the expected one. Recent approaches leverage Large Language Models (LLMs), trained on an enormous amount of data, to generate developer-like code and test cases. We investigate whether the LLM-generated test oracles capture the actual or expected software behaviour. We thus, conduct a controlled experiment to answer this question, by studying LLMs performance on two tasks, namely, test oracle classification and generation. The study includes developer-written and automatically generated test cases and oracles for 24 open-source Java repositories, and different well tested prompts. Our findings show that LLM-based test generation approaches are also prone on generating oracles that capture the actual program behaviour rather than the expected one. Moreover, LLMs are better at generating test oracles rather than classifying the correct ones, and can generate better test oracles when the code contains meaningful test or variable names. Finally, LLM-generated test oracles have higher fault detection potential than the Evosuite ones.

These notes focus on the minimization of convex functionals using first-order optimization methods, which are fundamental in many areas of applied mathematics and engineering. The primary goal of this document is to introduce and analyze the most classical first-order optimization algorithms. We aim to provide readers with both a practical and theoretical understanding in how and why these algorithms converge to minimizers of convex functions. The main algorithms covered in these notes include gradient descent, Forward-Backward splitting, Douglas-Rachford splitting, the Alternating Direction Method of Multipliers (ADMM), and Primal-Dual algorithms. All these algorithms fall into the class of first order methods, as they only involve gradients and subdifferentials, that are first order derivatives of the functions to optimize. For each method, we provide convergence theorems, with precise assumptions and conditions under which the convergence holds, accompanied by complete proofs. Beyond convex optimization, the final part of this manuscript extends the analysis to nonconvex problems, where we discuss the convergence behavior of these same first-order methods under broader assumptions. To contextualize the theory, we also include a selection of practical examples illustrating how these algorithms are applied in different image processing problems.

Flatback airfoils, characterized by a blunt trailing edge, are used at the root of large wind turbine blades. A low-drag pocket has recently been identified in the flow past these airfoils at high angles of attack, potentially offering opportunities for enhanced energy extraction. This study uses three-dimensional Detached Eddy Simulations (DES) combined with statistical and data-driven modal analysis techniques to explore the aerodynamics and coherent structures of a flatback airfoil in these conditions. Two angles of attack - one inside $\left(12^{\circ}\right)$ and one outside $\left(0^{\circ}\right)$ of the low-drag pocket - are examined more thoroughly. The spanwise correlation length of secondary instability is analyzed in terms of autocorrelation of the $\Gamma_{1}$ vortex identification criterion, while coherent structures were extracted via the multiscale Proper Orthogonal Decomposition (mPOD). The results show increased base pressure, BL thickness, vortex formation length, and more organized wake structures inside the low-drag regime. While the primary instability (B\'enard-von K\'arm\'an vortex street) dominates in both cases, the secondary instability is distinguishable only for the $12^{\circ}$ case and is identified as a Mode S$^{\prime}$ instability., Comment: 27 pages, 25 figures, submitted to Physical Review Fluids

In this paper, we present our proposed approach for active tracking to increase the autonomy of Unmanned Aerial Vehicles (UAVs) using event cameras, low-energy imaging sensors that offer significant advantages in speed and dynamic range. The proposed tracking controller is designed to respond to visual feedback from the mounted event sensor, adjusting the drone movements to follow the target. To leverage the full motion capabilities of a quadrotor and the unique properties of event sensors, we propose an end-to-end deep-reinforcement learning (DRL) framework that maps raw sensor data from event streams directly to control actions for the UAV. To learn an optimal policy under highly variable and challenging conditions, we opt for a simulation environment with domain randomization for effective transfer to real-world environments. We demonstrate the effectiveness of our approach through experiments in challenging scenarios, including fast-moving targets and changing lighting conditions, which result in improved generalization capabilities.

This paper addresses the modeling and attitude control of jumping quadrupeds in low-gravity environments. First, a convex decomposition procedure is presented to generate high-accuracy and low-cost collision geometries for quadrupeds performing agile maneuvers. A hierarchical control architecture is then investigated, separating torso orientation tracking from the generation of suitable, collision-free, corresponding leg motions. Nonlinear Model Predictive Controllers (NMPCs) are utilized in both layers of the controller. To compute the necessary leg motions, a torque allocation strategy is employed that leverages the symmetries of the system to avoid self-collisions and simplify the respective NMPC. To plan periodic trajectories online, a Finite State Machine (FSM)-based weight switching strategy is also used. The proposed controller is first evaluated in simulation, where 90 degree rotations in roll, pitch, and yaw are stabilized in 6.3, 2.4, and 5.5 seconds, respectively. The performance of the controller is further experimentally demonstrated by stabilizing constant and changing orientation references. Overall, this work provides a framework for the development of advanced model-based attitude controllers for jumping legged systems., Comment: 16 pages, 10 figures, to appear at the International Symposium of Robotics Research (ISRR) 2024. Paper site: https://michalispapadakis.github.io/mpc_olympus/

Active galactic nuclei (AGNs) are known to be variable across all wavelengths. Significant observational efforts have been invested in the last decade in studying their ultraviolet (UV) and optical variability. Long and densely sampled, multi-wavelength monitoring campaigns of numerous Seyfert galaxies have been conducted with the aim of determining the X-ray/UV/optical continuum time lags. Time-lag studies can be used to constrain theoretical models. The observed time lags can be explained by thermal reprocessing of the X-rays illuminating the accretion disc (known as the X-ray reverberation model). However, the observed light curves contain more information that can be used to further constrain physical models. Our primary objective is to investigate whether, in addition to time lags, the X-ray reverberation model can also explain the UV/optical variability amplitude of nearby Seyferts. To do this, we measured the excess variance of four sources (namely Mrk 509, NGC 4151, NGC 2617, and Mrk 142) as a function of wavelength using data from archival long, multi-wavelength campaigns with Swift, and ground-based telescopes. We also computed the model excess variance in the case of the X-ray reverberation model by determining the disc's transfer function and assuming a bending power law for the X-ray power spectrum. We tested the validity of the model by comparing the measured and model variances for a range of accretion rates and X-ray source heights. We conclude that the X-ray thermal reverberation model can fit both the continuum, UV/optical time lags, as well as the variance in these AGNs, for the same physical parameters. Our results suggest that the accretion disc is constant and that all the observed UV/optical variations, on timescales of days and up to a few weeks, can be fully explained by the variable X-rays as they illuminate the accretion disc., Comment: 16 pages, 10 figures, accepted for publication in Astronomy & Astrophysics

We study some representations of symmetric groups arising from a certain ideal in the coordinate ring of affine n-space. Our results give graded and representation-theoretic enhancements of sequence 337 of the Online Encyclopaedia of Integer Sequences, involving a symmetric version of the Bernoulli triangle., Comment: 8 pages. Main article by Nsibiet E. Udo, Praise Adeyemo and Balazs Szendroi; Appendix by Stavros Argyrios Papadakis

We present a multi-wavelength spectral study of NGC 4151 based on five epochs of simultaneous AstroSat observations in the near ultra-violet (NUV) to hard X-ray band ($\sim 0.005-80$ keV) during $2017 - 2018$. We derived the intrinsic accretion disk continuum after correcting for internal and Galactic extinction, contributions from broad and narrow line regions, and emission from the host galaxy. We found a bluer continuum at brighter UV flux possibly due to variations in the accretion disk continuum or the UV reddening. We estimated the intrinsic reddening, $E(B-V) \sim 0.4$, using high-resolution HST/STIS spectrum acquired in March 2000. We used thermal Comptonization, neutral and ionized absorption, and X-ray reflection to model the X-ray spectra. We obtained the X-ray absorbing neutral column varying between $N_H \sim 1.2-3.4 \times 10^{23} cm^{-2}$, which are $\sim 100$ times larger than that estimated from UV extinction, assuming the Galactic dust-to-gas ratio. To reconcile this discrepancy, we propose two plausible configurations of the obscurer: (a) a two-zone obscurer consisting of dust-free and dusty regions, divided by the sublimation radius, or (b) a two-phase obscurer consisting of clumpy, dense clouds embedded in a low-density medium, resulting in a scenario where a few dense clouds obscure the compact X-ray source substantially, while the bulk of UV emission arising from the extended accretion disk passes through the low-density medium. Furthermore, we find a positive correlation between X-ray absorption column and $NUV-FUV$ color and UV flux, indicative of enhanced winds possibly driven by the 'bluer-when-brighter' UV continuum., Comment: 21 pages, 22 figures, 6 tables, accepted for publication in ApJ

Thermal radiation is an abundant form of incoherent light. Generating coherent infrared light through incandescence promises a cheap alternative to the costly and epitaxially complex quantum cascade laser, however it remains a fundamental challenge. Previous approaches leveraged the spatial coherence of polaritonic excitations that occur in the thermal near-field, by diffracting them into the far-field zone via patterned micro- or nano-scatterers. This approach requires high-resolution lithography, is difficult to scale-up, and yields limited outcoupled radiation due to the intrinsically polarized nature of polaritons. We overcome these limitations and report coherent thermal emission through simple wave interference. We show that unpatterned, millimeter-scale, suspended nanomechanical membranes of SiC operate for both linear polarizations and exhibit antenna-like directionality without relying on the excitation of near-field polaritons. The ability to generate polarization-insensitive, narrowband and spatially coherent incandescent light without lithography at large scales paves the way towards democratizing thermal infrared technologies.

The ability to detect and engineer chirality plays a defining role in understanding nature, presenting itself in various phenomena ranging from the formation of DNA to enzymatic activities and digital encoding. Inducing a chiral response using nanophotonic practices typically entails lithographic efforts to create micro- and nano-structures that break mirror symmetry. Even with such configurations and despite the ubiquitousness of thermal radiation, generating chiral light at mid-infrared frequencies through thermal emission remains a challenge, since incandescence is by nature incoherent and unpolarized. In contrast to lithography-based approaches, we demonstrate that unpatterned twisted bilayers of in-plane anisotropic materials serve as a new material platform for intrinsic mid-infrared chirality engineering. Via absorption spectroscopy as well as direct thermal emission measurements, we report large circular dichroism and asymmetric circularly polarized thermal emission from exfoliated twisted bilayers of $\alpha$-molybdenum trioxide ($\alpha$-MoO$_3$). These findings highlight the potential of twisted layered van der Waals materials for polarization control and advanced thermal infrared technologies.

Recent works of the authors have demonstrated the usefulness of considering moduli spaces of Artinian reductions of a given ring when studying standard graded rings and their Lefschetz properties. This paper illuminates a key aspect of these works, the behaviour of the canonical module under deformations in this moduli space. We demonstrate that even when there is no natural geometry around, we can give a viewpoint that behaves like it, effectively constructing geometry out of nothing, giving interpretation to intersection numbers without cycles. Moreover, we explore some properties of this normalization., Comment: 30 pages

Our collective attention span is shortened by the flood of online information. With \textit{FarFetched}, we address the need for automated claim validation based on the aggregated evidence derived from multiple online news sources. We introduce an entity-centric reasoning framework in which latent connections between events, actions, or statements are revealed via entity mentions and represented in a graph database. Using entity linking and semantic similarity, we offer a way for collecting and combining information from diverse sources in order to generate evidence relevant to the user's claim. Then, we leverage textual entailment recognition to quantitatively determine whether this assertion is credible, based on the created evidence. Our approach tries to fill the gap in automated claim validation for less-resourced languages and is showcased on the Greek language, complemented by the training of relevant semantic textual similarity (STS) and natural language inference (NLI) models that are evaluated on translated versions of common benchmarks., Comment: DeepLo NAACL 2022

Mutation testing is a foundation approach in the software testing field, based on automatically seeded small syntactic changes, known as mutations. The question of how to generate high-utility mutations, to be used for testing purposes, forms a key challenge in mutation testing literature. Large Language Models (LLMs) have shown great potential in code-related tasks but their utility in mutation testing remains unexplored. To this end, we systematically investigate the performance of LLMs in generating effective mutations w.r.t. to their usability, fault detection potential, and relationship with real bugs. In particular, we perform a large-scale empirical study involving six LLMs, including both state-of-the-art open- and closed-source models, and 851 real bugs on two Java benchmarks (i.e., 605 bugs from 12 projects of Defects4J 2.0 and 246 bugs of ConDefects). We find that compared to existing approaches, LLMs generate more diverse mutations that are behaviorally closer to real bugs, which leads to approximately 19% higher fault detection than current approaches (i.e., 93% vs. 74%). Nevertheless, the mutants generated by LLMs have worse compilability rate, useless mutation rate, and equivalent mutation rate than those generated by rule-based approaches. This paper also examines alternative prompt engineering strategies and identifies the root causes of uncompilable mutations, providing insights for researchers to further enhance the performance of LLMs in mutation testing., Comment: 23 pages, 3 figures

Vlasiator is a space plasma simulation code which models near-Earth ion-kinetic dynamics in three spatial and three velocity dimensions. It is highly parallelized, modeling the Vlasov equation directly through the distribution function, discretized on a Cartesian grid, instead of the more common particle-in-cell approach. Modeling near-Earth space, plasma properties span several orders of magnitude in temperature, density, and magnetic field strength. In order to fit the required six-dimensional grids in memory, Vlasiator utilizes a sparse block-based velocity mesh, where chunks of velocity space are added or deleted based on the advection requirements of the Vlasov solver. In addition, the spatial mesh is adaptively refined through cell-based octree refinement. In this paper, we describe the design choices of porting Vlasiator to heterogeneous CPU/GPU architectures. We detail the memory management, algorithmic changes, and kernel construction as well as our unified codebase approach, resulting in portability to both NVIDIA and AMD hardware (CUDA and HIP languages, respectively). In particular, we showcase a highly parallel block adjustment approach allowing efficient re-ordering of a sparse velocity mesh. We detail pitfalls we have overcome and lay out a plan for optimization to facilitate future exascale simulations using multi-node GPU supercomputing., Comment: 31 pages, 12 figures, submitted to ASTRONUM 2024 - the 16th International Conference on Numerical Modeling of Space Plasma Flows

Class incremental learning consists in training discriminative models to classify an increasing number of classes over time. However, doing so using only the newly added class data leads to the known problem of catastrophic forgetting of the previous classes. Recently, dynamic deep learning architectures have been shown to exhibit a better stability-plasticity trade-off by dynamically adding new feature extractors to the model in order to learn new classes followed by a compression step to scale the model back to its original size, thus avoiding a growing number of parameters. In this context, we propose a new algorithm that enhances the compression of previous class knowledge by cutting and mixing patches of previous class samples with the new images during compression using our Rehearsal-CutMix method. We show that this new data augmentation reduces catastrophic forgetting by specifically targeting past class information and improving its compression. Extensive experiments performed on the CIFAR and ImageNet datasets under diverse incremental learning evaluation protocols demonstrate that our approach consistently outperforms the state-of-the-art . The code will be made available upon publication of our work.

Approaching thermodynamic limits in light harvesting requires enabling nonreciprocal thermal emission. The majority of previously reported nonreciprocal thermal emitters operate in reflection mode, following original proposals by M. Green and others. In these proposals, cascaded nonreciprocal junctions that re-direct each junction's emission towards a subsequent one are employed for efficient light-harvesting. Recently, simplified concepts have been proposed in solar photovoltaics and thermophotovoltaics that leverage the concept of tandem junctions to approach thermodynamic limits. In these simplified scenarios, polarization-independent nonreciprocal response in transmission mode is required. We propose a pattern-free heterostructure that enables such functionality, using a magneto-optical material embedded between two dissimilar dielectric layers., Comment: 9 pages, 6 figures

Context: Quasar variability has proven to be a powerful tool to constrain the properties of their inner engine and the accretion process onto supermassive black holes. Correlations between UV/optical variability and physical properties have been long studied with a plethora of different approaches and time-domain surveys, although the detailed picture is not yet clear. Aims: We analysed archival data from the SDSS Stripe-82 region to study how the quasar power spectral density (PSD) depends on the black hole mass, bolometric luminosity, accretion rate, redshift, and rest-frame wavelength. We developed a model-independent analysis framework that could be easily applied to upcoming large surveys such as the Legacy Survey of Space and Time (LSST). Methods: We used light curves of 8042 spectroscopically confirmed quasars, observed in at least six yearly seasons in five filters ugriz. We split the sample into bins of similar physical properties containing at least 50 sources, and we measured the ensemble PSD in each of them. Results: We find that a simple power law is a good fit to the power spectra in the frequency range explored. Variability does not depend on redshift at a fixed wavelength. Instead, both PSD amplitude and slope depend on the black hole mass, accretion rate, and rest-frame wavelength. We provide scaling relations to model the observed variability as a function of the physical properties, and discuss the possibility of a universal PSD shape for all quasars, where frequencies scale with the black hole mass, while normalization and slope(s) are fixed (at any given wavelength and accretion rate)., Comment: Accepted for publication in A&A

Recently, nearby active galactic nuclei (AGN) have been subject to long X-rays/UV/optical monitoring campaigns. These campaigns reveal a strong correlation between the various UV and optical bands, with time lags increasing with wavelength. In a series of papers, we demonstrated that a scenario in which a central X-ray source illuminates the accretion disc explains the observed correlations. However, some of the monitored AGN show low/moderate X-rays-UV correlations, which could challenge this scenario. In this paper, we study the broadband X-ray/UV/optical spectral energy distributions (SEDs) of NGC 5548, one of the most intensively monitored AGN. We aim to test if the X-ray illumination model explains the broadband spectral behaviour of the source, despite the moderate X-ray-UV/optical correlation. We model the broadband time-averaged SED, from the STORM monitoring campaign of the source, using the KYNSED model which assumes an X-ray illuminated disc. We assume that the accretion process powers the X-ray corona. We also model 15 time-resolved SEDs from the same campaign to check whether this scenario can account for the observed spectral variability. The proposed model describes well the time-averaged and the time-resolved SEDs of NGC 5548. In this scenario, the corona height, the X-ray photon index, and the power transferred to the corona all vary. This explains the variability behaviour at different wavelengths. The best-fit model is obtained for a non-rotating black hole accreting at a constant rate of 5% its Eddington limit. Since each of the variable parameters affects the observed flux in a particular way, the combined variability of all parameters explains the moderate X-ray-UV/optical correlation. The X-ray illuminated disc model provides a complete description of the behaviour of NGC 5548, explaining its broadband SEDs, time-lag spectrum, and its power spectral distribution., Comment: Accepted for publication in A&A

The explosion in the size and the complexity of the available Knowledge Graphs on the web has led to the need for efficient and effective methods for their understanding and exploration. Semantic summaries have recently emerged as methods to quickly explore and understand the contents of various sources. However in most cases they are static not incorporating user needs and preferences and cannot scale. In this paper we present iSummary a novel scalable approach for constructing personalized summaries. As the size and the complexity of the Knowledge Graphs for constructing personalized summaries prohibit efficient summary construction, in our approach we exploit query logs. The main idea behind our approach is to exploit knowledge captured in existing user queries for identifying the most interesting resources and linking them constructing as such highquality personalized summaries. We present an algorithm with theoretical guarantees on the summarys quality linear in the number of queries available in the query log. We evaluate our approach using three realworld datasets and several baselines showing that our approach dominates other methods in terms of both quality and efficiency.

Pre-trained code models lead the era of code intelligence. Many models have been designed with impressive performance recently. However, one important problem, data augmentation for code data that automatically helps developers prepare training data lacks study in the field of code learning. In this paper, we introduce a general data augmentation framework, GenCode, to enhance the training of code understanding models. GenCode follows a generation-and-selection paradigm to prepare useful training codes. Specifically, it uses code transformation techniques to generate new code candidates first and then selects important ones as the training data by importance metrics. To evaluate the effectiveness of GenCode with a general importance metric -- loss value, we conduct experiments on four code understanding tasks (e.g., code clone detection) and three pre-trained code models (e.g., CodeT5). Compared to the state-of-the-art (SOTA) code augmentation method, MixCode, GenCode produces code models with 2.92% higher accuracy and 4.90% robustness on average.

We study the X-ray power spectrum of Active Galactic Nuclei (AGN) to investigate whether Seyfert I and II power spectra are similar or not, whether the AGN variability depends on black hole mass and accretion rate, and to compare the AGN power spectra with the Galactic X-ray black hole binaries power-spectra. We used 14-195 keV band light curves from the 157th SWIFT/BAT hard X-ray survey and we computed the mean power spectrum and excess variance of AGN in narrow black-hole mass/luminosity bins. We fitted a power-law model to the AGN power spectra, and we investigated whether the power spectrum parameters and the excess variance depend on black hole mass, luminosity and accretion rate. The Seyfert I and Seyfert II power spectra are identical, in agreement with AGN unification models. The mean AGN X-ray power spectrum has the same, power-law like shape with a slope of -1 in all AGN, irrespective of their luminosity and BH mass. We do not detect any flattening to a slope of zero at frequencies as low as 10e-9 Hz. We detect an anti-correlation between the PSD amplitude and the accretion rate, similar to what has been seen in the past in the 2-10 keV band. This implies that the variability amplitude in AGN decreases with increasing accretion rate. The universal AGN power-spectrum is consistent with the mean, 2-9 keV band Cyg X-1 power spectrum in its soft state. The mean, low frequency AGN X-ray power spectrum is consistent with the extension of the mean, 0.01-25 Hz Cyg X-1 power spectrum to lower frequencies. The agreement between the AGN and the Cyg X-1 power spectrum (either in the soft or the hard state) over many decades in frequency indicates that the X-ray variability process is probably the same in all accreting objects, irrespective of the mass of the compact object. We plan to investigate this issue further in the near future., Comment: 13 pages, 10 figures. Accepted for publication in A&A

In this paper, narrowband emitters have been designed using particle swarm optimization (PSO) in the 10-20 {\mu}m infrared range. The device structure consists of an anisotropic {\alpha}-MoO3 layer combined with the one- and two-dimensional guided-mode resonance structures. Well-defined absorption lines are present in the reflection spectrum for both TE and TM polarizations, thereby yielding narrowband emissivity at desired wavelengths. The band structure of the designed emitters under TM polarization demonstrates distinct features unlike its TE counterpart. These features are attributed to the interaction between guided-mode resonances and phonon polaritons. The results are relevant for applications in active and passive photonic elements in mid- and long-wave IR bands.

RESTful APIs are popular web services, requiring documentation to ease their comprehension, reusability and testing practices. The OpenAPI Specification (OAS) is a widely adopted and machine-readable format used to document such APIs. However, manually documenting RESTful APIs is a time-consuming and error-prone task, resulting in unavailable, incomplete, or imprecise documentation. As RESTful API testing tools require an OpenAPI specification as input, insufficient or informal documentation hampers testing quality. Recently, Large Language Models (LLMs) have demonstrated exceptional abilities to automate tasks based on their colossal training data. Accordingly, such capabilities could be utilized to assist the documentation and testing process of RESTful APIs. In this paper, we present RESTSpecIT, the first automated RESTful API specification inference and black-box testing approach leveraging LLMs. The approach requires minimal user input compared to state-of-the-art RESTful API inference and testing tools; Given an API name and an LLM key, HTTP requests are generated and mutated with data returned by the LLM. By sending the requests to the API endpoint, HTTP responses can be analyzed for inference and testing purposes. RESTSpecIT utilizes an in-context prompt masking strategy, requiring no model fine-tuning. Our evaluation demonstrates that RESTSpecIT is capable of: (1) inferring specifications with 85.05% of GET routes and 81.05% of query parameters found on average, (2) discovering undocumented and valid routes and parameters, and (3) uncovering server errors in RESTful APIs. Inferred specifications can also be used as testing tool inputs.

Plug-and-Play (PnP) algorithms are a class of iterative algorithms that address image inverse problems by combining a physical model and a deep neural network for regularization. Even if they produce impressive image restoration results, these algorithms rely on a non-standard use of a denoiser on images that are less and less noisy along the iterations, which contrasts with recent algorithms based on Diffusion Models (DM), where the denoiser is applied only on re-noised images. We propose a new PnP framework, called Stochastic deNOising REgularization (SNORE), which applies the denoiser only on images with noise of the adequate level. It is based on an explicit stochastic regularization, which leads to a stochastic gradient descent algorithm to solve ill-posed inverse problems. A convergence analysis of this algorithm and its annealing extension is provided. Experimentally, we prove that SNORE is competitive with respect to state-of-the-art methods on deblurring and inpainting tasks, both quantitatively and qualitatively.

The black hole mass MBH is crucial in constraining the growth of supermassive BHs within their host galaxies. Since direct measurements of MBH with dynamical methods are restricted to a limited number of nearly quiescent nearby galaxies and a small minority of active galactic nuclei (AGN), we must rely on indirect methods. In this work, we utilize an unbiased, volume-limited, hard X-ray selected sample of AGN to compare the reliability of some commonly used indirect methods, emphasising those that can be applied to obscured AGN. Based on a subsample of AGN with MBH determined via dynamical methods, our study suggests that X-ray based techniques, such as the scaling method and the one based on the variability measured through the excess variance, are in good agreement with the dynamical methods. On the other hand, the M-sigma correlation based on inactive galaxies tends to systematically overestimate MBH, regardless of the level of obscuration. We provide a correcting factor that produces an acceptable agreement with dynamical values and can be used to quickly correct the MBH computed with this method. We also derive an alternative M-sigma correlation based on this unbiased sample of AGN with a slope considerably shallower than the ones obtained using inactive galaxies, suggesting that the latter correlation may not be appropriate to compute the MBH in AGN. Finally, we find that no quick fix can be applied to correct the MBH obtained from the fundamental plane of black hole activity, casting doubts on the reliability of this method., Comment: 15 pages, 9 figures, 3 tables, Accepted for publication in MNRAS

A thermophotovoltaic (TPV) energy converter harnesses thermal photons emitted by a hot body and converts them to electricity. When the radiative heat exchange between the emitter and PV cell is spectrally monochromatic, the TPV system can approach the Carnot thermodynamic efficiency limit. Nonetheless, this occurs at the expense of vanishing extracted electrical power density. Conversely, a spectrally broadband radiative heat exchange between the emitter and the cell yields maximal TPV power density at the expense of low efficiency. By leveraging hot-carriers as a means to mitigate thermalization losses within the cell, we demonstrate that one can alleviate this trade-off between power density and efficiency. Via detailed balance analysis, we show analytically that one can reach near-Carnot conversion efficiencies close to the maximum power point, which is unattainable with conventional TPV systems. We derive analytical relations between intrinsic device parameters and performance metrics, which serve as design rules for hot-carrier-based TPV systems., Comment: 10 pages, 7 figures

We study the connection between the variations in the far ultra-violet (FUV), near ultra-violet (NUV) and X-ray band emission from NGC 4051 using 4-days long AstroSat observations performed during 5-9 June 2016. NGC 4051 showed rapid variability in all three bands with the strongest variability amplitude in the X-ray band ($F_{var} \sim 37\%$) and much weaker variability in the UV bands ($F_{var} \sim 3 - 5\%$). Cross-correlation analysis performed using Interpolated cross-correlation Functions (ICCF) and Discrete cross-correlation Functions (DCF) revealed a robust correlation ($\sim 0.75$) between the UV and X-ray light curves. The variations in the X-ray band are found to lead those in the FUV and NUV bands by $\sim 7.4{\rm~ks}$ and $\sim 24.2{\rm~ks}$, respectively. The UV lags favour the thermal disc reprocessing model. The FUV and NUV bands are strongly correlated ($\sim 0.9$) and the variations in the FUV band lead those in the NUV band by $\sim 13{\rm~ks}$. Comparison of the UV lags found using the AstroSat observations with those reported earlier and the theoretical model for thermal reverberation time-lag suggests a possible change in either the geometry of the accretion disc/corona or the height of the corona., Comment: 8 pages, 5 figures, Accepted for publication in MNRAS

In this work, we present new proofs of convergence for Plug-and-Play (PnP) algorithms. PnP methods are efficient iterative algorithms for solving image inverse problems where regularization is performed by plugging a pre-trained denoiser in a proximal algorithm, such as Proximal Gradient Descent (PGD) or Douglas-Rachford Splitting (DRS). Recent research has explored convergence by incorporating a denoiser that writes exactly as a proximal operator. However, the corresponding PnP algorithm has then to be run with stepsize equal to $1$. The stepsize condition for nonconvex convergence of the proximal algorithm in use then translates to restrictive conditions on the regularization parameter of the inverse problem. This can severely degrade the restoration capacity of the algorithm. In this paper, we present two remedies for this limitation. First, we provide a novel convergence proof for PnP-DRS that does not impose any restrictions on the regularization parameter. Second, we examine a relaxed version of the PGD algorithm that converges across a broader range of regularization parameters. Our experimental study, conducted on deblurring and super-resolution experiments, demonstrate that both of these solutions enhance the accuracy of image restoration., Comment: arXiv admin note: substantial text overlap with arXiv:2301.13731

We consider the effect of multiple stochastic parameters on the time-average quantities of chaotic systems. We employ the recently proposed \cite{Kantarakias_Papadakis_2023} sensitivity-enhanced generalized polynomial chaos expansion, se-gPC, to compute efficiently this effect. se-gPC is an extension of gPC expansion, enriched with the sensitivity of the time-averaged quantities with respect to the stochastic variables. To compute these sensitivities, the adjoint of the shadowing operator is derived in the frequency domain. Coupling the adjoint operator with gPC provides an efficient uncertainty quantification (UQ) algorithm which, in its simplest form, has computational cost that is independent of the number of random variables. The method is applied to the Kuramoto-Sivashinsky equation and is found to produce results that match very well with Monte-Carlo simulations. The efficiency of the proposed method significantly outperforms sparse-grid approaches, like Smolyak Quadrature. These properties make the method suitable for application to other dynamical systems with many stochastic parameters.

We develop Self2Seg, a self-supervised method for the joint segmentation and denoising of a single image. To this end, we combine the advantages of variational segmentation with self-supervised deep learning. One major benefit of our method lies in the fact, that in contrast to data-driven methods, where huge amounts of labeled samples are necessary, Self2Seg segments an image into meaningful regions without any training database. Moreover, we demonstrate that self-supervised denoising itself is significantly improved through the region-specific learning of Self2Seg. Therefore, we introduce a novel self-supervised energy functional in which denoising and segmentation are coupled in a way that both tasks benefit from each other. We propose a unified optimisation strategy and numerically show that for noisy microscopy images our proposed joint approach outperforms its sequential counterpart as well as alternative methods focused purely on denoising or segmentation.

This paper serves as a foundational step towards the development of a linguistically motivated and technically relevant evaluation suite for Greek NLP. We initiate this endeavor by introducing four expert-verified evaluation tasks, specifically targeted at natural language inference, word sense disambiguation (through example comparison or sense selection) and metaphor detection. More than language-adapted replicas of existing tasks, we contribute two innovations which will resonate with the broader resource and evaluation community. Firstly, our inference dataset is the first of its kind, marking not just \textit{one}, but rather \textit{all} possible inference labels, accounting for possible shifts due to e.g. ambiguity or polysemy. Secondly, we demonstrate a cost-efficient method to obtain datasets for under-resourced languages. Using ChatGPT as a language-neutral parser, we transform the Dictionary of Standard Modern Greek into a structured format, from which we derive the other three tasks through simple projections. Alongside each task, we conduct experiments using currently available state of the art machinery. Our experimental baselines affirm the challenging nature of our tasks and highlight the need for expedited progress in order for the Greek NLP ecosystem to keep pace with contemporary mainstream research., Comment: EACL 2023 (Findings)

In this paper, we present an updated version of our model (KYNXiltr) which considers thermal reverberation of a standard Novikov-Thorne accretion disc illuminated by an X-ray point-like source. Previously, the model considered only two cases of black hole spins, and assumed a colour correction factor $f_{\rm col} = 2.4$. Now, we extend the model to any spin value and colour correction. In addition, we consider two scenarios of powering the X-ray corona, either via accretion, or external to the accretion disc. We use KYNXiltr to fit the observed time lags obtained from intense monitoring of four local Seyfert galaxies (NGC 5548, NGC 4395, Mrk 817, and Fairall 9). We consider various combinations of black hole spin, colour correction, corona height, and fraction of accretion power transferred to the corona. The model fits well the overall time-lags spectrum in these sources (for a large parameter space). For NGC 4593 only, we detect a significant excess of delays in the U-band. The contribution of the diffuse BLR emission in the time-lags spectrum of this source is significant. It is possible to reduce the large best-fitting parameter space by combining the results with additional information, such as the observed Eddington ratio and average X-ray luminosity. We also provide an update to the analytic expression provided by Kammoun et al., for an X-ray source that is not powered by the accretion process, which can be used for any value of colour correction, and for two values of the black hole spin (0 and 0.998)., Comment: Accepted for publication in MNRAS

Much research on Machine Learning testing relies on empirical studies that evaluate and show their potential. However, in this context empirical results are sensitive to a number of parameters that can adversely impact the results of the experiments and potentially lead to wrong conclusions (Type I errors, i.e., incorrectly rejecting the Null Hypothesis). To this end, we survey the related literature and identify 10 commonly adopted empirical evaluation hazards that may significantly impact experimental results. We then perform a sensitivity analysis on 30 influential studies that were published in top-tier SE venues, against our hazard set and demonstrate their criticality. Our findings indicate that all 10 hazards we identify have the potential to invalidate experimental findings, such as those made by the related literature, and should be handled properly. Going a step further, we propose a point set of 10 good empirical practices that has the potential to mitigate the impact of the hazards. We believe our work forms the first step towards raising awareness of the common pitfalls and good practices within the software engineering community and hopefully contribute towards setting particular expectations for empirical research in the field of deep learning testing.