Search

We consider the simultaneous optimal transportation of measures, where the target marginal is not necessarily fixed. For this problem, we prove the existence of a solution for completely regular spaces and investigate the structure of the discrete problem. We establish a connection between the Monge problem and the Kantorovich problem by showing that their functionals are equal and that the solutions coincide in Euclidean space., Comment: 11 pages

We discuss a prediction of the solar activity on a short time-scale applying the method based on a combination of a nonlinear mean-field dynamo model and the artificial neural network. The artificial neural network which serves as a correction scheme for the forecast, uses the currently available observational data (e.g., the 13 month running average of the observed solar sunspot numbers) and the dynamo model output. The nonlinear mean-field $\alpha\,\Omega$ dynamo produces the large-scale magnetic flux which is redistributed by negative effective magnetic pressure instability (NEMPI) producing sunspots and active regions. The nonlinear mean-field dynamo model includes algebraic nonlinearity (caused by the feedback of the growing magnetic field on the plasma motion) and dynamic nonlinearities (related to the dynamics of the magnetic helicity of small-scale magnetic field). We compare the forecast errors with a horizon of 1, 6, 12 and 18 months, for different forecast methods, with the same corrections on the current monthly observations. Our forecast is in good agreement with the observed solar activity, the forecast error is almost stably small over short-medium ranges of forecasting windows. Despite a strong level of chaotic component in the solar magnetic activity we present quantitative evidence that the solar activity on a short range can be stably well predicted, by the joint use of the physically based model with the neural network. This result may have an immediate practical implementation for predictions of various phenomena of solar activity and other astrophysical processes, so may be of interest to a broad community., Comment: 12 pages, 4 figures, 1 table

To balance the quality and inference cost of a Foundation Model (FM, such as large language models (LLMs)) powered software, people often opt to train a routing model that routes requests to FMs with different sizes and capabilities. Existing routing models rely on learning the optimal routing decision from carefully curated data, require complex computations to be updated, and do not consider the potential evolution of weaker FMs. In this paper, we propose Real-time Adaptive Routing (RAR), an approach to continuously adapt FM routing decisions while using guided in-context learning to enhance the capabilities of weaker FM. The goal is to reduce reliance on stronger, more expensive FMs. We evaluate our approach on different subsets of the popular MMLU benchmark. Over time, our approach routes 50.2% fewer requests to computationally expensive models while maintaining around 90.5% of the general response quality. In addition, the guides generated from stronger models have shown intra-domain generalization and led to a better quality of responses compared to an equivalent approach with a standalone weaker FM.

A main goal in developing video-compression algorithms is to enhance human-perceived visual quality while maintaining file size. But modern video-analysis efforts such as detection and recognition, which are integral to video surveillance and autonomous vehicles, involve so much data that they necessitate machine-vision processing with minimal human intervention. In such cases, the video codec must be optimized for machine vision. This paper explores the effects of compression on detection and recognition algorithms (objects, faces, and license plates) and introduces novel full-reference image/video-quality metrics for each task, tailored to machine vision. Experimental results indicate our proposed metrics correlate better with the machine-vision results for the respective tasks than do existing image/video-quality metrics., Comment: 16 pages, 10 figures

Two-dimensional materials offering ultrafast photoresponse suffer from low intrinsic absorbance, especially in the mid-infrared wavelength range. Challenges in 2d material doping further complicate the creation of light-sensitive $p-n$ junctions. Here, we experimentally demonstrate a graphene-based infrared detector with simultaneously enhanced absorption and strong structural asymmetry enabling zero-bias photocurrent. A key element for those properties is an asymmetric singular metasurface (ASMS) atop graphene with keen metal wedges providing singular enhancement of local absorbance. The ASMS geometry predefines extra device functionalities. The structures with connected metallic wedges demonstrate polarization ratios up to 200 in a broad range of carrier densities at a wavelength of 8.6 $\mu$m. The structures with isolated wedges display gate-controlled switching between polarization-discerning and polarization-stable photoresponse, a highly desirable yet scarce property for polarized imaging.

The far-from-equilibrium dynamics of spatial Polyakov loop correlations, which provide gauge-invariant observables akin to effective particle numbers for gluon plasmas, are investigated within real-time $\mathrm{SU}(N_c)$ lattice gauge theory at weak couplings and large gluon occupations. The momentum zero mode of these correlations reveals the dynamic formation of a condensate, while at nonzero momenta, energy is transported toward the ultraviolet. We demonstrate that the non-zero momentum dynamics is well described by a direct cascade in terms of gauge-invariant Polyakov loop excitations, exhibiting self-similar prescaling indicative of a nonthermal attractor. This behavior can be analytically understood through perturbation theory for the Polyakov loop correlations and the established dynamics of gauge field correlations. We perform simulations for both $\mathrm{SU}(2)$ and $\mathrm{SU}(3)$ gauge groups, providing further consistency checks on the $N_c$-dependence of perturbative expectations. No evidence of an inverse cascade toward lower momenta is found for momenta above the electric screening scale., Comment: 12 pages, 8 figures

Generating high output powers while achieving narrow line single mode lasing are often mutual exclusive properties of commercial laser diodes. For this reason, efficient and scalable amplification of narrow line laser light is still a major driving point in modern laser system designs. Commonly, injection locking of high-power semiconductor laser diodes are used for this purpose. However, for many laser diodes it is very challenging to achieve stable operation of the injection locked state due to a complex interplay of non-linearities and thermal effects. Different approaches of active or passive stabilization usually require a large overhead of optical and electrical equipment and are not generally applicable. In our work we present a passive stabilization scheme, that is generally applicable, technically easy to implement and extremely cost-effective. It is based on the externally synchronized automatic acquisition of the optimal injection state. Central to our simple but powerful scheme is the management of thermalization effects during lock acquisition. By periodical relocking, spectrally pure amplified light is maintained in a quasi-CW manner over long timescales. We characterize the performance of our method for laser diodes amplifying 671 nm light and demonstrate the general applicability by confirming the method to work also for laser diodes at 401 nm, 461 nm and 689 nm. Our scheme enables the scaled operation of injection locks, even in cascaded setups, for the distributed amplification of single frequency laser light.

As foundation models (FMs) play an increasingly prominent role in complex software systems, such as FM-powered agentic software (i.e., Agentware), they introduce significant challenges for developers regarding observability. Unlike traditional software, agents operate autonomously, using extensive data and opaque implicit reasoning, making it difficult to observe and understand their behavior during runtime, especially when they take unexpected actions or encounter errors. In this paper, we highlight the limitations of traditional operational observability in the context of FM-powered software, and introduce cognitive observability as a new type of required observability that has emerged for such innovative systems. We then propose a novel framework that provides cognitive observability into the implicit reasoning processes of agents (a.k.a. reasoning observability), and demonstrate the effectiveness of our framework in boosting the debuggability of Agentware and, in turn, the abilities of an Agentware through a case study on AutoCodeRover, a cuttingedge Agentware for autonomous program improvement.

The need of atmospheric information with a higher spatial and temporal resolution drives the development of small satellites and satellite constellations to complement satellite flagship missions. Since optical systems are a main contributor to the satellite size, these are the prime candidate for their miniaturization. We present here a novel optical system where the complete spectrometer part of the optical system is compressed in one flat optical element. The element consists of an array of photonic crystals which is directly placed on a detector. The photonic crystals act as optical filters with a tunable spectral transmission response. From the integrated optical signals per filter and the atmosphere model, greenhouse gas concentrations are obtained using computational inversion. We present in this article the instrument concept, the manufacturing and measurement of the photonic crystals, methods for the filter array optimization, and discuss the predicted retrieval performance for the detection of methane and carbon dioxide., Comment: ICSO 2024

Cycle packing is a fundamental problem in optimization, graph theory, and algorithms. Motivated by recent advancements in finding vertex-disjoint paths between a specified set of vertices that either minimize the total length of the paths [Bj\"orklund, Husfeldt, ICALP 2014; Mari, Mukherjee, Pilipczuk, and Sankowski, SODA 2024] or request the paths to be shortest [Lochet, SODA 2021], we consider the following cycle packing problems: Min-Sum Cycle Packing and Shortest Cycle Packing. In Min-Sum Cycle Packing, we try to find, in a weighted undirected graph, $k$ vertex-disjoint cycles of minimum total weight. Our first main result is an algorithm that, for any fixed $k$, solves the problem in polynomial time. We complement this result by establishing the W[1]-hardness of Min-Sum Cycle Packing parameterized by $k$. The same results hold for the version of the problem where the task is to find $k$ edge-disjoint cycles. Our second main result concerns Shortest Cycle Packing, which is a special case of Min-Sum Cycle Packing that asks to find a packing of $k$ shortest cycles in a graph. We prove this problem to be fixed-parameter tractable (FPT) when parameterized by $k$ on weighted planar graphs. We also obtain a polynomial kernel for the edge-disjoint variant of the problem on planar graphs. Deciding whether Min-Sum Cycle Packing is FPT on planar graphs and whether Shortest Cycle Packing is FPT on general graphs remain challenging open questions.

Accurate characterization of ferroelectric properties in polymer piezoelectrics is critical for optimizing the performance of flexible and wearable ultrasound transducers, such as screen-printed PVDF devices. Standard charge measurement techniques, like the Sawyer-Tower circuit, often fall short when applied to ferroelectric polymers due to low-frequency leakage. In this work, we present PEtra, an open-source and versatile piezoelectric loop tracer. PEtra employs a transimpedance amplifier (LMP7721, TI) to convert picoampere-level currents into measurable voltages, covering a frequency range of 0.1 Hz to 5 Hz for a gain setting of 10^7 V/A, and 0.1 Hz to 200 Hz for gain settings between 10^3 V/A to 10^6 V/A (10-fold increments). We demonstrate through simulations and experimental validations that PEtra achieves a sensitivity down to 2 pA, effectively addressing the limitations of traditional charge measurement methods. Compared to the Sawyer-Tower circuit, PEtra directly amplifies currents without the need for a reference capacitor. As a result, it is less susceptible to leakage and can operate at lower frequencies, improving measurement accuracy and reliability. PEtra's design is fully open source, offering researchers and engineers a versatile tool to drive advancements in flexible PVDF transducer technology.

New line lists for four isotopologues of nickel monohydride, $^{58}$NiH, $^{60}$NiH, $^{62}$NiH, and $^{58}$NiD are presented covering the wavenumber range $<10000$ cm$^{-1}$ ($\lambda > 1$ $\mu$m), $J$ up to 37.5 for transitions within and between the three lowest-lying electronic states, ${X}\,^{2}\Delta$, ${W}\,^{2}\Pi$, and ${V}\,^{2}\Sigma^{+}$. The line lists are applicable for temperatures up to 5000 K. The line lists calculations are based on a recent empirical NiH spectroscopic model [Havalyova et al. J. Quant. Spectrosc. Radiat. Transf., 272, 107800, (2021)] which is adapted for the variational nuclear-motion code Duo. The model consists of potential energy curves, spin-orbit coupling curves, electronic angular momentum curves, spin-rotation coupling curves, $\Lambda$-doubling correction curve for $^2\Pi$ states and Born-Oppenheimer breakdown (BOB) rotational correction curves. New ab initio dipole moment curves, scaled to match the experimental dipole moment of the ground state, are used to compute Einstein A coefficients. The BYOT line lists are included in the ExoMol database at www.exomol.com.

We develop a perturbation theory for surfaces confining photons and massive particles in static spherically symmetric spacetimes in terms of two parameters: the mass-to-energy ratio and the deviation of metric functions from a given form, e.g., the Schwarzschild solution. Expansions of the gravitational shadow radius in terms of these parameters are constructed up to the second order. The metric expansion in terms of the Schwarzschild mass-to-radius ratio is then reconstructed. Explicit analytical examples of non-standard black hole metrics are considered as an illustration. In some cases perturbative results demonstrate good accuracy even for non-small deviations., Comment: 26 pages, 5 figures, revtex4

The vulnerability of artificial neural networks to adversarial perturbations in the black-box setting is widely studied in the literature. The majority of attack methods to construct these perturbations suffer from an impractically large number of queries required to find an adversarial example. In this work, we focus on knowledge distillation as an approach to conduct transfer-based black-box adversarial attacks and propose an iterative training of the surrogate model on an expanding dataset. This work is the first, to our knowledge, to provide provable guarantees on the success of knowledge distillation-based attack on classification neural networks: we prove that if the student model has enough learning capabilities, the attack on the teacher model is guaranteed to be found within the finite number of distillation iterations.

In countries with a growing number of elderly and a shrinking workforce, one of which is Russia, it becomes impossible to maintain a solidary pension system and a need to switch to a more stable funded system appears. This paper analyzes various scenarios of Russia's transition to such a system. This is the first study on the Russian economy in which an Overlapping Generations Model is used to simulate the pension transition. It is demonstrated that in the long term, the transition to a funded system slightly reduces the welfare of pensioners, and during the transition, the situation of pensioners deteriorates strongly. However, it is also important to emphasize that the transition imposes a heavy burden on all generations living during the reform, they are forced to consume less and greatly change their savings, while also often starting to work more. Such conclusions are made concerning average population cohorts, and the results may not be the same for different groups of individuals within these cohorts. In different scenarios, the pension system transition can cause both economic growth and economic recession, as well as a corresponding increase or decrease in wages and consumption., Comment: Comments: 23 pages, in Russian

This work addresses the critical challenge of optimal filter selection for a novel trace gas measurement device. This device uses photonic crystal filters to retrieve trace gas concentrations prone to photon and read noise. The filter selection directly influences accuracy and precision of the gas retrieval and therefore is a crucial performance driver. We formulate the problem as a stochastic combinatorial optimization problem and develop a simulator mimicking gas retrieval with noise. The objective function for selecting filters reducing retrieval error is minimized by the employed metaheuristics, that represent various families of optimizers. We aim to improve the found top-performing algorithms using our novel distance-driven extensions, that employ metrics on the space of filter selections. This leads to a novel adaptation of the UMDA algorithm, we call UMDA-U-PLS-Dist, equipped with one of the proposed distance metrics as the most efficient and robust solver among the considered ones. Analysis of filter sets produced by this method reveals that filters with relatively smooth transmission profiles but containing high contrast improve the device performance. Moreover, the top-performing obtained solution shows significant improvement compared to the baseline.

Hadron-to-two-hadron transition generalized parton distributions (GPDs) extend the concept of hadron-to-resonance transition GPDs and provide a unified description of non-diagonal hard exclusive reactions in the generalized Bjorken limit. We present the formalism for the case of spinless hadrons addressing the non-diagonal deeply virtual Compton scattering $\gamma^*\pi\to\gamma\pi\pi$ in terms of $\pi\to\pi\pi$ transition GPDs, which generalize GPDs for $\pi \to f_0, \, \rho, \, f_2, \, \cdots$ transitions. We work out the basic properties of $\pi\to\pi\pi$ transition GPDs and establish the soft pion theorems at the $2\pi$ production threshold. We construct the partial wave expansion of $\pi\to\pi\pi$ transition GPDs in the two-pion decay angles and employ the dispersive approach to constrain $\pi\to\pi\pi$ transition GPDs in terms of $\pi\pi$-scattering phases with help of the Omn\`{e}s representation. We estimate the $e^-\pi^+ \to e^- \gamma \pi^+ \pi^0$ cross section in the kinematics of the JLab@12 GeV incorporating the isolated $\rho(770)$ resonance state and work out the angular distributions of the cross section, specifying the observables sensitive to the polarization states of the produced $\rho(770)$ resonance., Comment: 45 pages, 15 figures

Autonomous navigation of a mobile robot is a challenging task which requires ability of mapping, localization, path planning and path following. Conventional mapping methods build a dense metric map like an occupancy grid, which is affected by odometry error accumulation and consumes a lot of memory and computations in large environments. Another approach to mapping is the usage of topological properties, e.g. adjacency of locations in the environment. Topological maps are less prone to odometry error accumulation and high resources consumption, and also enable fast path planning because of the graph sparsity. Based on this idea, we proposed NavTopo - a full navigation pipeline based on topological map and two-level path planning. The pipeline localizes in the graph by matching neural network descriptors and 2D projections of the input point clouds, which significantly reduces memory consumption compared to metric and topological point cloud-based approaches. We test our approach in a large indoor photo-relaistic simulated environment and compare it to a metric map-based approach based on popular metric mapping method RTAB-MAP. The experimental results show that our topological approach significantly outperforms the metric one in terms of performance, keeping proper navigational efficiency., Comment: This paper is published in proceedings of the 9th International Conference "Interactive Collaborative Robotics" (ICR 2024)

DNA-Encoded Libraries (DEL) are combinatorial small molecule libraries that offer an efficient way to characterize diverse chemical spaces. Selection experiments using DELs are pivotal to drug discovery efforts, enabling high-throughput screens for hit finding. However, limited availability of public DEL datasets hinders the advancement of computational techniques designed to process such data. To bridge this gap, we present KinDEL, one of the first large, publicly available DEL datasets on two kinases: Mitogen-Activated Protein Kinase 14 (MAPK14) and Discoidin Domain Receptor Tyrosine Kinase 1 (DDR1). Interest in this data modality is growing due to its ability to generate extensive supervised chemical data that densely samples around select molecular structures. Demonstrating one such application of the data, we benchmark different machine learning techniques to develop predictive models for hit identification; in particular, we highlight recent structure-based probabilistic approaches. Finally, we provide biophysical assay data, both on- and off-DNA, to validate our models on a smaller subset of molecules. Data and code for our benchmarks can be found at: https://github.com/insitro/kindel.

Thanks to their record high refractive index and giant optical anisotropy, van der Waals (vdW) materials have accelerated the development of nanophotonics. However, traditional high refractive index materials, such as titanium dioxide (TiO2), still dominate in the most important visible range. This is due to the current lack of transparent vdW materials across the entire visible spectrum. In this context, we propose that germanium disulfide (GeS2) could offer a significant breakthrough. With its high refractive index, negligible losses, and biaxial optical anisotropy across the whole visible range, GeS2 has the potential to complement TiO2 and close the application gap of vdW materials in the visible spectrum. The addition of GeS2 could have a profound impact on the design of van der Waals nanophotonic circuits for any operation wavelength from ultraviolet to infrared, emphasizing the significance of the potential impact of GeS2 on the field of nanophotonics., Comment: 15 pages, 4 figures

Room temperature ionic liquids show great promise as electrolytes in various technological applications, such as energy storage or electrotunable lubrication. These applications are particularly intriguing due to the specific behavior of ionic liquids in nanoconfinement. While previous research has been focused on optimizing the required characteristics through the selection of electrolyte properties, the contribution of confining material properties in these systems has been largely overlooked. In this Review, we provide constructive analysis of recent developments related to the description of surface material properties impact on the ionic liquid behavior in the confinement and propose potential ways for further investigations in this direction. Although the presented advances reveal the importance of surface material properties in the application processes with confined ionic liquids, there are still a lot of issues that should be thoroughly investigated in future. We believe that this review will significantly contribute to the development of new approaches with material properties consideration for confined ionic liquid research., Comment: 21 pages, 6 figures

In this brief paper, we analyze space weather events that occurred on May 11 and 12, 2024, from the perspective of an operational space weather center that provides advisories for civil aviation. One of the key metrics monitored by the center is the radiation dose rate at operational flight altitudes. A model implemented by the center provides the dose rate in real time. The model showed that dangerous levels were momentarily exceeded just above the usual 30,000 feet level during the events. This paper highlights differences in models used by various space weather centers, emphasizing the need for harmonization., Comment: 7 pages, 4 figures

Rare event sampling in dynamical systems is a fundamental problem arising in the natural sciences, which poses significant computational challenges due to an exponentially large space of trajectories. For settings where the dynamical system of interest follows a Brownian motion with known drift, the question of conditioning the process to reach a given endpoint or desired rare event is definitively answered by Doob's h-transform. However, the naive estimation of this transform is infeasible, as it requires simulating sufficiently many forward trajectories to estimate rare event probabilities. In this work, we propose a variational formulation of Doob's h-transform as an optimization problem over trajectories between a given initial point and the desired ending point. To solve this optimization, we propose a simulation-free training objective with a model parameterization that imposes the desired boundary conditions by design. Our approach significantly reduces the search space over trajectories and avoids expensive trajectory simulation and inefficient importance sampling estimators which are required in existing methods. We demonstrate the ability of our method to find feasible transition paths on real-world molecular simulation and protein folding tasks., Comment: Accepted as Spotlight at Conference on Neural Information Processing Systems (NeurIPS 2024); Alanine dipeptide results updated after fixing unphysical parameterization and energy computation

Due to the development of facial manipulation techniques in recent years deepfake detection in video stream became an important problem for face biometrics, brand monitoring or online video conferencing solutions. In case of a biometric authentication, if you replace a real datastream with a deepfake, you can bypass a liveness detection system. Using a deepfake in a video conference, you can penetrate into a private meeting. Deepfakes of victims or public figures can also be used by fraudsters for blackmailing, extorsion and financial fraud. Therefore, the task of detecting deepfakes is relevant to ensuring privacy and security. In existing approaches to a deepfake detection their performance deteriorates when multiple faces are present in a video simultaneously or when there are other objects erroneously classified as faces. In our research we propose to use geometric-fakeness features (GFF) that characterize a dynamic degree of a face presence in a video and its per-frame deepfake scores. To analyze temporal inconsistencies in GFFs between the frames we train a complex deep learning model that outputs a final deepfake prediction. We employ our approach to analyze videos with multiple faces that are simultaneously present in a video. Such videos often occur in practice e.g., in an online video conference. In this case, real faces appearing in a frame together with a deepfake face will significantly affect a deepfake detection and our approach allows to counter this problem. Through extensive experiments we demonstrate that our approach outperforms current state-of-the-art methods on popular benchmark datasets such as FaceForensics++, DFDC, Celeb-DF and WildDeepFake. The proposed approach remains accurate when trained to detect multiple different deepfake generation techniques., Comment: 10 pages, 6 figures

Robots' behavior and performance are determined both by hardware and software. The design process of robotic systems is a complex journey that involves multiple phases. Throughout this process, the aim is to tackle various criteria simultaneously, even though they often contradict each other. The ultimate goal is to uncover the optimal solution that resolves these conflicting factors. Generative, computation or automatic designs are the paradigms aimed at accelerating the whole design process. Within this paper we propose a design methodology to generate linkage mechanisms for robots with morphological computation. We use a graph grammar and a heuristic search algorithm to create robot mechanism graphs that are converted into simulation models for testing the design output. To verify the design methodology we have applied it to a relatively simple quasi-static problem of object grasping. We found a way to automatically design an underactuated tendon-driven gripper that can grasp a wide range of objects. This is possible because of its structure, not because of sophisticated planning or learning.

Among interesting applications of cold atoms, quantum simulations attract a lot of attention. In this context, rare-earth ultracold atoms are particularly appealing for such simulators due to their numerous Fano-Feshbach resonances and magnetic dipole moments in the ground state. Creating a quantum gas microscope requires a large optical access that may be achieved using transport of atoms between separate vacuum volumes. We demonstrate that in case of the transport of magnetic atoms the magnetic field can be directly measured and adjusted to reduce additional losses after the transport therefore increasing the efficiency of subsequent evaporation cooling. This approach allows to transfer over 85% of the atoms from the main chamber to the scientific chamber, located 38 cm away with moderate laser power of 26 W without atomic polarization decay.

This article addresses the question of reporting a lower confidence band (LCB) for optimal welfare in a policy learning problem. A straightforward procedure inverts a one-sided t-test based on an efficient estimator of the optimal welfare. We show that under empirically relevant data-generating processes, this procedure can be dominated by an LCB corresponding to suboptimal welfare, with the average difference of the order N-1/2. We relate the first-order dominance result to a lack of uniformity in the margin assumption, a standard sufficient condition for debiased inference on the optimal welfare ensuring that the first-best policy is well-separated from the suboptimal ones. Finally, we show that inverting the existing tests from the moment inequality literature produces LCBs that are robust to the non-uniqueness of the optimal policy and easy to compute. We find that this approach performs well empirically in the context of the National JTPA study.

Metals in the diffuse, ionized gas at the boundary between the Milky Way's interstellar medium (ISM) and circumgalactic medium (CGM), known as the disk-halo interface (DHI), are valuable tracers of the feedback processes that drive the Galactic fountain. However, metallicity measurements in this region are challenging due to obscuration by the Milky Way ISM and uncertain ionization corrections that affect the total hydrogen column density. In this work, we constrain the ionization corrections to neutral hydrogen column densities using precisely measured electron column densities from the dispersion measure of pulsars that lie in the same globular clusters as UV-bright targets with high-resolution absorption spectroscopy. We address the blending of absorption lines with the ISM by jointly fitting Voigt profiles to all absorption components. We present our metallicity estimates for the DHI of the Milky Way based on detailed photoionization modeling to the absorption from ionized metal lines and ionization-corrected total hydrogen columns. Generally, the gas clouds show a large scatter in metallicity, ranging between $0.04-3.2\ Z_{\odot}$, implying that the DHI consists of a mixture of gaseous structures having multiple origins. We estimate the inflow and outflow timescales of the DHI ionized clouds to be $6 - 35$ Myr. We report the detection of an infalling cloud with super-solar metallicity that suggests a Galactic fountain mechanism, whereas at least one low-metallicity outflowing cloud ($Z < 0.1\ Z_{\odot}$) poses a challenge for Galactic fountain and feedback models., Comment: 28 pages, 15 figures, accepted for publication in ApJ

In this paper, we address the challenge of speech enhancement in real-world recordings, which often contain various forms of distortion, such as background noise, reverberation, and microphone artifacts. We revisit the use of Generative Adversarial Networks (GANs) for speech enhancement and theoretically show that GANs are naturally inclined to seek the point of maximum density within the conditional clean speech distribution, which, as we argue, is essential for the speech enhancement task. We study various feature extractors for perceptual loss to facilitate the stability of adversarial training, developing a methodology for probing the structure of the feature space. This leads us to integrate WavLM-based perceptual loss into MS-STFT adversarial training pipeline, creating an effective and stable training procedure for the speech enhancement model. The resulting speech enhancement model, which we refer to as FINALLY, builds upon the HiFi++ architecture, augmented with a WavLM encoder and a novel training pipeline. Empirical results on various datasets confirm our model's ability to produce clear, high-quality speech at 48 kHz, achieving state-of-the-art performance in the field of speech enhancement. Demo page: https://samsunglabs.github.io/FINALLY-page, Comment: Accepted to NeurIPS 2024

In Clique Cover, given a graph $G$ and an integer $k$, the task is to partition the vertices of $G$ into $k$ cliques. Clique Cover on unit ball graphs has a natural interpretation as a clustering problem, where the objective function is the maximum diameter of a cluster. Many classical NP-hard problems are known to admit $2^{O(n^{(1 - 1/d)})}$-time algorithms on unit ball graphs in $\mathbb{R}^d$ [de Berg et al., SIAM J. Comp 2018]. A notable exception is the Maximum Clique problem, which admits a polynomial-time algorithm on unit disk graphs and a subexponential algorithm on unit ball graphs in $\mathbb{R}^3$, but no subexponential algorithm on unit ball graphs in dimensions 4 or larger, assuming the ETH [Bonamy et al., JACM 2021]. In this work, we show that Clique Cover also suffers from a "curse of dimensionality", albeit in a significantly different way compared to Maximum Clique. We present a $2^{O(\sqrt{n})}$-time algorithm for unit disk graphs and argue that it is tight under the ETH. On the other hand, we show that Clique Cover does not admit a $2^{o(n)}$-time algorithm on unit ball graphs in dimension $5$, unless the ETH fails.

We extend the formalism of tri-vector deformations to the full SL(5) exceptional field theory with no truncation assumed thus covering 11D backgrounds of any form. We derive explicit transformation rules for 11D supergravity component fields and prove that these generate solutions given the same algebraic conditions hold: generalized Yang-Baxter equation and the unimodularity condition.

V745 Sco is a Galactic symbiotic recurrent nova with nova eruptions in 1937, 1989 and 2014. We study the behavior of V745 Sco at radio wavelengths (0.6-37,GHz), covering both its 1989 and 2014 eruptions and informed by optical, X-ray, and $\gamma$-ray data. The radio light curves are synchrotron-dominated. Surprisingly, compared to expectations for synchrotron emission from explosive transients such as radio supernovae, the light curves spanning 0.6-37 GHz all peak around the same time ($\sim$18-26 days after eruption) and with similar flux densities (5-9 mJy).We model the synchrotron light curves as interaction of the nova ejecta with the red giant wind, but find that simple spherically symmetric models with wind-like circumstellar material (CSM) cannot explain the radio light curve. Instead, we conclude that the shock suddenly breaks out of a dense CSM absorbing screen around 20 days after eruption, and then expands into a relatively low density wind ($\dot{M}_{out} \approx 10^{-9}-10^{-8}$ M$_{\odot}$ yr$^{-1}$ for $v_w = 10$ km s$^{-1}$) out to $\sim$1 year post-eruption. The dense, close-in CSM may be an equatorial density enhancement or a more spherical red giant wind with $\dot{M}_{in} \approx [5-10] \times 10^{-7}$ M$_{\odot}$ yr$^{-1}$, truncated beyond several $\times 10^{14}$ cm. The outer lower-density CSM would not be visible in typical radio observations of Type Ia supernovae: V745 Sco cannot be ruled out as a Type Ia progenitor based on CSM constraints alone.Complementary constraints from the free-free radio optical depth and the synchrotron luminosity imply the shock is efficient at accelerating relativistic electrons and amplifying magnetic fields, with $\epsilon_e$ and $\epsilon_B \approx 0.01-0.1$., Comment: 21 pages, 20 figures

We study electromagnetic radiation by electrically charged fermions embedded in a rotating medium in an external magnetic field. We compute the dependence of the radiation intensity on the angular velocity $\Omega$ of the rotating medium for fermion polarizations along and opposite the magnetic field direction in intense subcritical fields. The polarization dependence of the photon radiation results in the Sokolov--Ternov effect -- the radiative polarization of fermions. We study the dependence of the degree of polarziation on $\Omega$. We found that rotation significantly changes the degree of polarization. We show that the rotating quark-gluon plasma acquires a finite magnetic moment that exhibits complex dependence on $\Omega$., Comment: 18 pages

The rapid and accurate detection of biochemical compositions in fish is a crucial real-world task that facilitates optimal utilization and extraction of high-value products in the seafood industry. Raman spectroscopy provides a promising solution for quickly and non-destructively analyzing the biochemical composition of fish by associating Raman spectra with biochemical reference data using machine learning regression models. This paper investigates different regression models to address this task and proposes a new design of Convolutional Neural Networks (CNNs) for jointly predicting water, protein, and lipids yield. To the best of our knowledge, we are the first to conduct a successful study employing CNNs to analyze the biochemical composition of fish based on a very small Raman spectroscopic dataset. Our approach combines a tailored CNN architecture with the comprehensive data preparation procedure, effectively mitigating the challenges posed by extreme data scarcity. The results demonstrate that our CNN can significantly outperform two state-of-the-art CNN models and multiple traditional machine learning models, paving the way for accurate and automated analysis of fish biochemical composition.

Aperiodic crystals are the intermediates between strictly periodic crystalline matter and amorphous solids. The lack of translational symmetry combined with intrinsic long-range order endows aperiodic crystals with unique physical characteristics, while at the same time dramatically enriching the spectrum and localization properties. Here, we demonstrate exciton-polariton condensation in a two-dimensional Penrose tiling with $C_{10}$ rotational symmetry - the first signature of quasicrystalline order in a quantum fluid of light. We identify a regime, wherein near-perfect delocalization and synchronization of a quantum fluid of light occurs at mesoscopic length-scales extending beyond 100x the healing length and the size of each individual condensate. Realizing long-range order in fully reconfigurable aperiodic crystals of nonlinear, and open-dissipative quantum fluids, lays the foundations for testing a broad range of universality classes of continuous phase transitions beyond the limits of mathematically verifiable models in regular lattices.

The emergent axion excitations in the quark-gluon plasma are generated by the sphaleron transitions. Their interactions with the photon and gluon fields are governed by the axion electro- and chromodynamics respectively. We discuss the effect of the emergent axion on the photon production and energy loss in the quark-gluon plasma. In particular, we review the Chiral Cherenkov effect, and the impact of the chiral anomaly on bremsstrahlung., Comment: Review for International Journal of Modern Physics E

Cascade multiplication is a common technique to enhance the sensitivity of photon detectors. In this study, we demonstrate novel cascade-amplified superconducting detectors utilizing arrays of Josephson junctions. The mutual coupling between junctions induces avalanche-like switching of multiple junctions upon photon absorption, leading to cascade amplification of the readout voltage. We present two prototypes featuring either low-Tc linear Nb/NbxSi1-x/Nb arrays, or high-Tc stacked intrinsic Josephson junctions. Both devices exhibit clear antenna effects in microwave directivities, indicating good impedance matching and absorption efficiency. The combination of high absorption efficiency and large cascade amplification has the potential to produce broadband THz sensors with sensitivity exceeding 10^13 V/W., Comment: 11 pages, 6 figures

Modern realities and trends in learning require more and more generalization ability of models, which leads to an increase in both models and training sample size. It is already difficult to solve such tasks in a single device mode. This is the reason why distributed and federated learning approaches are becoming more popular every day. Distributed computing involves communication between devices, which requires solving two key problems: efficiency and privacy. One of the most well-known approaches to combat communication costs is to exploit the similarity of local data. Both Hessian similarity and homogeneous gradients have been studied in the literature, but separately. In this paper, we combine both of these assumptions in analyzing a new method that incorporates the ideas of using data similarity and clients sampling. Moreover, to address privacy concerns, we apply the technique of additional noise and analyze its impact on the convergence of the proposed method. The theory is confirmed by training on real datasets., Comment: 25 pages, 15 figures, 4 appendices

Feeding the growing human population sustainably amidst climate change is one of the most important challenges in the 21st century. Current practices often lead to the overuse of agronomic inputs, such as synthetic fertilizers and water, resulting in environmental contamination and diminishing returns on crop productivity. The complexity of agricultural systems, involving plant-environment interactions and human management, presents significant scientific and technical challenges for developing sustainable practices. Addressing these challenges necessitates transdisciplinary research, involving intense collaboration among fields such as plant science, engineering, computer science, and social sciences. Here, we present five case studies from two research centers demonstrating successful transdisciplinary approaches toward more sustainable water and fertilizer use. These case studies span multiple scales. Starting from whole-plant signaling, we explore how reporter plants can transform our understanding of plant communication and enable efficient application of water and fertilizers. We then show how new fertilizer technologies could increase the availability of phosphorus in the soil. To accelerate advancements in breeding new cultivars, we discuss robotic technologies for high-throughput plant screening in different environments at a population scale. At the ecosystem scale, we investigate phosphorus recovery from aquatic systems and methods to minimize phosphorus leaching. Finally, as agricultural outputs affect all people, we show how to integrate stakeholder perspectives and needs into the research. With these case studies, we hope to encourage the scientific community to adopt transdisciplinary research and promote cross-training among biologists, engineers, and social scientists to drive discovery and innovation in advancing sustainable agricultural systems.

We present the high-precision result for the zero-jettiness soft function at next-to-next-to-next-to-leading order (N3LO) in perturbative QCD. At this perturbative order, the soft function is the last missing ingredient required for the computation of a hadronic colour singlet production or a colour singlet decay into two jets using the zero-jettiness variable as the slicing parameter. Furthermore, the knowledge of the N3LO soft function enables the re-summed description of the thrust distribution in the process $e^+ e^- \to \textrm{hadrons}$ through next-to-next-to-next-to-leading logarithmic order, which is important for the extraction of the strong coupling constant using this shape variable. On the methodological side, the complexity of the zero-jettiness variable forced us to develop a new semi-analytic method for phase-space integration in the presence of constraints parameterized through Heaviside functions which, hopefully, will be useful for further development of the $N$-jettiness slicing scheme., Comment: 6 pages, attached results in computer readable form

In the emerging field of computational gastronomy, aligning culinary practices with scientifically supported nutritional goals is increasingly important. This study explores how large language models (LLMs) can be applied to optimize ingredient substitutions in recipes, specifically to enhance the phytochemical content of meals. Phytochemicals are bioactive compounds found in plants, which, based on preclinical studies, may offer potential health benefits. We fine-tuned models, including OpenAI's GPT-3.5, DaVinci, and Meta's TinyLlama, using an ingredient substitution dataset. These models were used to predict substitutions that enhance phytochemical content and create a corresponding enriched recipe dataset. Our approach improved Hit@1 accuracy on ingredient substitution tasks, from the baseline 34.53 plus-minus 0.10% to 38.03 plus-minus 0.28% on the original GISMo dataset, and from 40.24 plus-minus 0.36% to 54.46 plus-minus 0.29% on a refined version of the same dataset. These substitutions led to the creation of 1,951 phytochemically enriched ingredient pairings and 1,639 unique recipes. While this approach demonstrates potential in optimizing ingredient substitutions, caution must be taken when drawing conclusions about health benefits, as the claims are based on preclinical evidence. Future work should include clinical validation and broader datasets to further evaluate the nutritional impact of these substitutions. This research represents a step forward in using AI to promote healthier eating practices, providing potential pathways for integrating computational methods with nutritional science., Comment: 15 pages

Recent theoretical studies of carrier-carrier scattering in degenerate two-dimensional systems have revealed radically different relaxation times for odd and even angular harmonics of distribution function. This theoretical concept, dubbed as 'tomographic Fermi liquid', is yet challenging to test with dc electrical measurements as electron scattering weakly affects the electrical resistivity. Here, we show that linewidth and amplitude of electromagnetic absorption at the multiple harmonics of the cyclotron resonance carries all necessary information to test the tomographic Fermi liquid hypothesis. Namely, the height and inverse width of $m$-th order cyclotron resonance ($m \ge 2$) is proportional to the lifetime of $m$-th angular harmonic of electron distribution function $\tau_m$, if probed at wavelengths exceeding the cyclotron radius $R_c$. Measurements of high-order cyclotron resonance at short wavelengths order of $R_c$ also enable a direct determination of all lifetimes $\tau_m$ from a simple linear system of equations that we hereby derive. Extraction of cyclotron resonance lifetimes from an experiment on terahertz photoconductivity in graphene shows that third-order resonance is systematically narrower than second-order one, supporting the prediction of tomographic Fermi liquid hypothesis.