Search

In contrast to problems of interference in (exogenous) treatments, models of interference in unit-specific (endogenous) outcomes do not usually produce a reduced-form representation where outcomes depend on other units' treatment status only at a short network distance, or only through a known exposure mapping. This remains true if the structural mechanism depends on outcomes of peers only at a short network distance, or through a known exposure mapping. In this paper, we first define causal estimands that are identified and estimable from a single experiment on the network under minimal assumptions on the structure of interference, and which represent average partial causal responses which generally vary with other global features of the realized assignment. Under a fixed-population, design-based approach, we show unbiasedness, consistency and asymptotic normality for inverse-probability weighting (IPW) estimators for those causal parameters from a randomized experiment on a single network. We also analyze more closely the case of marginal interventions in a model of equilibrium with smooth response functions where we can recover LATE-type weighted averages of derivatives of those response functions. Under additional structural assumptions, these "agnostic" causal estimands can be combined to recover model parameters, but also retain their less restrictive causal interpretation.

The goal of protein fitness optimization is to discover new protein variants with enhanced fitness for a given use. The vast search space and the sparsely populated fitness landscape, along with the discrete nature of protein sequences, pose significant challenges when trying to determine the gradient towards configurations with higher fitness. We introduce Variational Latent Generative Protein Optimization (VLGPO), a variational perspective on fitness optimization. Our method embeds protein sequences in a continuous latent space to enable efficient sampling from the fitness distribution and combines a (learned) flow matching prior over sequence mutations with a fitness predictor to guide optimization towards sequences with high fitness. VLGPO achieves state-of-the-art results on two different protein benchmarks of varying complexity. Moreover, the variational design with explicit prior and likelihood functions offers a flexible plug-and-play framework that can be easily customized to suit various protein design tasks.

For the exploration of the phase diagram of QCD, effective Polyakov loop theories derived from lattice QCD provide a valuable tool in the heavy quark mass regime. Using mean field approximations these theories are evaluated in the high and low temperature regimes at finite baryon chemical potential. The resulting phase diagram is discussed., Comment: 9 pages, 3 figures, contribution to the 41st International Symposium on Lattice Field Theory (Lattice 2024), 28 July - 3 August 2024, Liverpool, UK

We demonstrate how neural networks can drive mathematical discovery through a case study of the Hadwiger-Nelson problem, a long-standing open problem from discrete geometry and combinatorics about coloring the plane avoiding monochromatic unit-distance pairs. Using neural networks as approximators, we reformulate this mixed discrete-continuous geometric coloring problem as an optimization task with a probabilistic, differentiable loss function. This enables gradient-based exploration of admissible configurations that most significantly led to the discovery of two novel six-colorings, providing the first improvements in thirty years to the off-diagonal variant of the original problem (Mundinger et al., 2024a). Here, we establish the underlying machine learning approach used to obtain these results and demonstrate its broader applicability through additional results and numerical insights., Comment: 8 pages main paper, 10 pages references and appendix, 17 figures, 1 table

We introduce a novel method for generating 360{\deg} panoramas from text prompts or images. Our approach leverages recent advances in 3D generation by employing multi-view diffusion models to jointly synthesize the six faces of a cubemap. Unlike previous methods that rely on processing equirectangular projections or autoregressive generation, our method treats each face as a standard perspective image, simplifying the generation process and enabling the use of existing multi-view diffusion models. We demonstrate that these models can be adapted to produce high-quality cubemaps without requiring correspondence-aware attention layers. Our model allows for fine-grained text control, generates high resolution panorama images and generalizes well beyond its training set, whilst achieving state-of-the-art results, both qualitatively and quantitatively. Project page: https://cubediff.github.io/, Comment: Accepted at ICLR 2025

Exploration of the Universe requires communication with Earth, either on a direct path or through a cascade of proximate celestial bodies. Microwaves have traditionally been used for space communication, but electromagnetic waves of higher frequencies, such as in the optical domain, will enable probing farther in space due to their considerably lower diffraction loss. At a given data rate, the ultimate limit to point-to-point optical communication is determined by the received signal power and the photon information efficiency. The latter measures the number of information bits extracted per photon incident on a detector. As distances across space and, consequently, path loss increases, the system that can achieve the highest photon information efficiency will determine the longest distance at which communication is possible. We report here an experimental demonstration of optical detection at a record photon information efficiency of 14.5 bits per incident photon, or 17.8 bits per detected photon, after 87.5 dB of attenuation. Expressed in terms of energy per bit, this corresponds to 8.84 zeptojoules per bit, or 0.069 photons per bit at 1550 nm. To our knowledge, this is the highest photon information efficiency or lowest energy per bit detection system ever demonstrated at optical frequencies. Such a sensitive detection system holds promise for a wide range of applications.

Hallucinations of deep neural models are amongst key challenges in automatic speech recognition (ASR). In this paper, we investigate hallucinations of the Whisper ASR model induced by non-speech audio segments present during inference. By inducting hallucinations with various types of sounds, we show that there exists a set of hallucinations that appear frequently. We then study hallucinations caused by the augmentation of speech with such sounds. Finally, we describe the creation of a bag of hallucinations (BoH) that allows to remove the effect of hallucinations through the post-processing of text transcriptions. The results of our experiments show that such post-processing is capable of reducing word error rate (WER) and acts as a good safeguard against problematic hallucinations., Comment: Accepted for IEEE ICASSP 2025

This paper presents LiFT, a lightweight, fully quantized 3D object detection algorithm for LiDAR data, optimized for real-time inference on FPGA platforms. Through an in-depth analysis of FPGA-specific limitations, we identify a set of FPGA-induced constraints that shape the algorithm's design. These include a computational complexity limit of 30 GMACs (billion multiply-accumulate operations), INT8 quantization for weights and activations, 2D cell-based processing instead of 3D voxels, and minimal use of skip connections. To meet these constraints while maximizing performance, LiFT combines novel mechanisms with state-of-the-art techniques such as reparameterizable convolutions and fully sparse architecture. Key innovations include the Dual-bound Pillar Feature Net, which boosts performance without increasing complexity, and an efficient scheme for INT8 quantization of input features. With a computational cost of just 20.73 GMACs, LiFT stands out as one of the few algorithms targeting minimal-complexity 3D object detection. Among comparable methods, LiFT ranks first, achieving an mAP of 51.84% and an NDS of 61.01% on the challenging NuScenes validation dataset. The code will be available at https://github.com/vision-agh/lift., Comment: The paper has been accepted for the DASIP 2025 workshop in conjunction with the HiPEAC 2025 conference in Barcelona

Electricity distribution cable networks suffer from incomplete and unbalanced data, hindering the effectiveness of machine learning models for predictive maintenance and reliability evaluation. Features such as the installation date of the cables are frequently missing. To address data scarcity, this study investigates the application of Variational Autoencoders (VAEs) for data enrichment, synthetic data generation, imbalanced data handling, and outlier detection. Based on a proof-of-concept case study for Denmark, targeting the imputation of missing age information in cable network asset registers, the analysis underlines the potential of generative models to support data-driven maintenance. However, the study also highlights several areas for improvement, including enhanced feature importance analysis, incorporating network characteristics and external features, and handling biases in missing data. Future initiatives should expand the application of VAEs by incorporating semi-supervised learning, advanced sampling techniques, and additional distribution grid elements, including low-voltage networks, into the analysis.

The torus in Active Galactic Nuclei (AGN) is a complex dynamical structure of gas and dust. It is thought to be composed of an equatorial dusty disk and a polar dusty wind launched by radiation pressure. However, this picture is based on studies of moderately accreting AGN. Models suggest that the disk/wind structure will change with specific accretion rate. Here we examine the wind launching region in two high accretion rate objects, I Zw 1 (super-Eddington) and H0557-385 (high-Eddington), using high spatial resolution interferometric observations in the $K$-band from VLTI/GRAVITY and $LM$-bands VLTI/MATISSE. We recover wavelength-dependent sizes of the dust emission using a Gaussian and power law fit to the visibilities. Both objects are partially resolved and have radial sizes in the $KLM$-bands between 0.3 - 1.5 mas, with no signs of elongation. Combining our measurements with VLTI/MIDI $N$-band data gives a full multi-wavelength picture of the dust structure. We find that in H0557-385, the dust sizes between $3.5-8\:\mu\mathrm{m}$ are independent of the wavelength, roughly constant at $3-10$ sublimation radii. We argue that this indicates a direct view of the wind launching region and, together with an absence of polar elongation, this implies that any wind would be launched in a preferentially equatorial direction or blown out by strong radiation pressure. The size-wavelength relation for both objects shows a preferentially disky equatorial dust distribution. We conclude that there is strong evidence that the Eddington ratio shapes the inner dust structure, most notably the wind-launching region and wind direction., Comment: 16 pages, 15 figures, 5 tables. Accepted for publication in MNRAS

This article presents a Non-negative Tensor Factorization based method for sound source separation from Ambisonic microphone signals. The proposed method enables the use of prior knowledge about the Directions-of-Arrival (DOAs) of the sources, incorporated through a constraint on the Spatial Covariance Matrix (SCM) within a Maximum a Posteriori (MAP) framework. Specifically, this article presents a detailed derivation of four algorithms that are based on two types of cost functions, namely the squared Euclidean distance and the Itakura-Saito divergence, which are then combined with two prior probability distributions on the SCM, that is the Wishart and the Inverse Wishart. The experimental evaluation of the baseline Maximum Likelihood (ML) and the proposed MAP methods is primarily based on first-order Ambisonic recordings, using four different source signal datasets, three with musical pieces and one containing speech utterances. We consider under-determined, determined, as well as over-determined scenarios by separating two, four and six sound sources, respectively. Furthermore, we evaluate the proposed algorithms for different spherical harmonic orders and at different reverberation time levels, as well as in non-ideal prior knowledge conditions, for increasingly more corrupted DOAs. Overall, in comparison with beamforming and a state-of-the-art separation technique, as well as the baseline ML methods, the proposed MAP approach offers superior separation performance in a variety of scenarios, as shown by the analysis of the experimental evaluation results, in terms of the standard objective separation measures, such as the SDR, ISR, SIR and SAR.

This manuscript studies the Gaussian approximation of the coordinate-wise maximum of self-normalized statistics in high-dimensional settings. We derive an explicit Berry-Esseen bound under weak assumptions on the absolute moments. When the third absolute moment is finite, our bound scales as $\log^{5/4}(d)/n^{1/8}$ where $n$ is the sample size and $d$ is the dimension. Hence, our bound tends to zero as long as $\log(d)=o(n^{1/10})$. Our results on self-normalized statistics represent substantial advancements, as such a bound has not been previously available in the high-dimensional central limit theorem (CLT) literature.

Disaster events occur around the world and cause significant damage to human life and property. Earth observation (EO) data enables rapid and comprehensive building damage assessment (BDA), an essential capability in the aftermath of a disaster to reduce human casualties and to inform disaster relief efforts. Recent research focuses on the development of AI models to achieve accurate mapping of unseen disaster events, mostly using optical EO data. However, solutions based on optical data are limited to clear skies and daylight hours, preventing a prompt response to disasters. Integrating multimodal (MM) EO data, particularly the combination of optical and SAR imagery, makes it possible to provide all-weather, day-and-night disaster responses. Despite this potential, the development of robust multimodal AI models has been constrained by the lack of suitable benchmark datasets. In this paper, we present a BDA dataset using veRy-hIGH-resoluTion optical and SAR imagery (BRIGHT) to support AI-based all-weather disaster response. To the best of our knowledge, BRIGHT is the first open-access, globally distributed, event-diverse MM dataset specifically curated to support AI-based disaster response. It covers five types of natural disasters and two types of man-made disasters across 12 regions worldwide, with a particular focus on developing countries where external assistance is most needed. The optical and SAR imagery in BRIGHT, with a spatial resolution between 0.3-1 meters, provides detailed representations of individual buildings, making it ideal for precise BDA. In our experiments, we have tested seven advanced AI models trained with our BRIGHT to validate the transferability and robustness. The dataset and code are available at https://github.com/ChenHongruixuan/BRIGHT. BRIGHT also serves as the official dataset for the 2025 IEEE GRSS Data Fusion Contest.

The aim of CeNTREX (Cold Molecule Nuclear Time-Reversal Experiment) is to search for time-reversal symmetry violation in the thallium nucleus, by measuring the Schiff moment of $^{205}$Tl in the polar molecule thallium fluoride (TlF). CeNTREX uses a cryogenic beam of TlF with a rotational temperature of 6.3(2) K. This results in population spread over dozens of rotational and hyperfine sublevels of TlF, while only a single level is useful for the Schiff moment measurement. Here we present a protocol for cooling the rotational and hyperfine degrees of freedom in the CeNTREX beam, transferring the majority of the Boltzmann distribution into a single rotational and hyperfine sublevel by using a single ultraviolet laser and a pair of microwave beams. We achieve a factor of $20.1(4)$ gain in the population of the $J=0$, $F=0$ hyperfine sublevel of the TlF ground state., Comment: 9 pages, 7 figures

Calving front position variation of marine-terminating glaciers is an indicator of ice mass loss and a crucial parameter in numerical glacier models. Deep Learning (DL) systems can automatically extract this position from Synthetic Aperture Radar (SAR) imagery, enabling continuous, weather- and illumination-independent, large-scale monitoring. This study presents the first comparison of DL systems on a common calving front benchmark dataset. A multi-annotator study with ten annotators is performed to contrast the best-performing DL system against human performance. The best DL model's outputs deviate 221 m on average, while the average deviation of the human annotators is 38 m. This significant difference shows that current DL systems do not yet match human performance and that further research is needed to enable fully automated monitoring of glacier calving fronts. The study of Vision Transformers, foundation models, and the inclusion and processing strategy of more information are identified as avenues for future research.

We show that Rufus Bowen's Problem 32 on the classification of symbolic systems with the specification property does not admit a solution that would use concrete invariants. To this end, we construct a class of symbolic systems with the specification property and show that the conjugacy relation on this class is too complicated to admit such a classification. More generally, we gauge the complexity of the classification problem for symbolic systems with the specification property. Along the way, we also provide answers to two questions related to the classification of pointed systems with the specification property: to a question of Ding and Gu related to the complexity of the classification of pointed Cantor systems with the specification property and to a question of Bruin and Vejnar related to the complexity of the classification of pointed Hilbert cube systems with the specification property.

The matter power spectrum, $P(k)$, is one of the fundamental quantities in the study of large-scale structure in cosmology. Here, we study its small-scale asymptotic limit, and show that for cold dark matter in $d$ spatial dimensions, $P(k)$ has a universal $k^{-d}$ asymptotic scaling with the wave-number $k$, for $k \gg k_{\rm nl}$, where $k_{\rm nl}^{-1}$ denotes the length scale at which non-linearities in gravitational interactions become important. We propose a theoretical explanation for this scaling, based on a non-perturbative analysis of the system's phase-space structure. Gravitational collapse is shown to drive a turbulent phase-space flow of the quadratic Casimir invariant, where the linear and non-linear time scales are balanced, and this balance dictates the $k$ dependence of the power spectrum. A parallel is drawn to Batchelor turbulence in hydrodynamics, where large scales mix smaller ones via tidal interactions. The $k^{-d}$ scaling is also derived by expressing $P(k)$ as a phase-space integral in the framework of kinetic field theory, which is analysed by the saddle-point method; the dominant critical points of this integral are precisely those where the time scales are balanced. The coldness of the dark-matter distribution function -- its non-vanishing only on a $d$-dimensional sub-manifold of phase-space -- underpins both approaches. The theory is accompanied by $1\mathrm{D}$ Vlasov--Poisson simulations, which confirm it., Comment: Submitted, comments welcome

Community detection, also known as graph partitioning, is a well-known NP-hard combinatorial optimization problem with applications in diverse fields such as complex network theory, transportation, and smart power grids. The problem's solution space grows drastically with the number of vertices and subgroups, making efficient algorithms crucial. In recent years, quantum computing has emerged as a promising approach to tackling NP-hard problems. This study explores the use of a quantum-inspired algorithm, Simulated Bifurcation (SB), for community detection. Modularity is employed as both the objective function and a metric to evaluate the solutions. The community detection problem is formulated as a Quadratic Unconstrained Binary Optimization (QUBO) problem, enabling seamless integration with the SB algorithm. Experimental results demonstrate that SB effectively identifies community structures in benchmark networks such as Zachary's Karate Club and the IEEE 33-bus system. Remarkably, SB achieved the highest modularity, matching the performance of Fujitsu's Digital Annealer, while surpassing results obtained from two quantum machines, D-Wave and IBM. These findings highlight the potential of Simulated Bifurcation as a powerful tool for solving community detection problems., Comment: 7 pages, 4 figures, 2 tables

By incorporating contributions from both the (chromo)electric scale $gT$ and (chromo)magnetic scale $g^2T$, we establish spectral sum rules of quarks for strongly coupled QCD that respect Fermi-Dirac statistics as required by quantum mechanics. In sharp contrast to QED and weakly coupled QCD whose spectral functions consist of discontinuous zero-dimensional (poles) and one-dimensional (branch cuts) non-analytic contributions from real energy $p_0 \in \mathbb{R}$, the derived spectral function for strongly coupled quarks features continuous but non-analytic contributions from complex energy $p_0 \in \mathbb{C}$ that are two-dimensional in nature. In light of the novel sum rules, we uncover an intrinsic QCD transition between a three-mode phase at small coupling and a one-mode phase at large coupling. The transition is induced by the magnetic scale that generates a massless hydro-like mode with the genuine non-Abelian feature of positivity violation and serving as the Goldstone mode of the Lorentz symmetry breaking. The thermal mass serves as an order parameter of the transition and vanishes at large coupling in line with phenomenological predictions from Dyson-Schwinger equations and gauge/gravity duality. This result provides novel insights into the mechanism of the QCD deconfinement transition., Comment: 13 pages, 10 figures

In the Felsenkeller shallow-underground site, protected from cosmic muons by a 45 m thick rock overburden, a research laboratory including a 5 MV Pelletron ion accelerator and a number of radioactivity-measurement setups is located. The laboratory and its installations are described in detail. The background radiation has been studied, finding suppression factors of 40 for cosmic-ray muons, 200 for ambient neutrons, and 100 for the background in germanium $\gamma$-ray detectors. Using an additional active muon veto, typically the background is just twice as high as in very deep underground laboratories. The properties of the accelerator including its external and internal ion sources and beam line are given. For the radioactivity counting setup, detection limits in the 10$^{-4}$ Bq range have been obtained. Practical aspects for the usage of the laboratory by outside scientific users are discussed., Comment: Submitted to Eur. Phys. J. A

Plasma turbulence in the edge of magnetic confinement devices is customarily treated as full-$f$ due to large fluctuations. For computational efficiency, field-aligned coordinates are employed, separating the magnetic field into equilibrium $B_0$ and delta-f perturbations which are handled by the magnetic flutter operators. Evolving the full-$f$ pressure with delta-$f$ magnetic perturbations can cause inconsistency since the latter contain background components such as the Shafranov shift, which are actually parts of the equilibrium magnetic field. Such background components ($B_s$) contained in the magnetic perturbations undermine the field-aligned numerics when treated as flutter: errors arise if $B_s/B_0\ll l_\perp/h_\parallel$ is not satisfied, with the perpendicular turbulence scale $l_\perp$ and the parallel grid distance $h_\parallel$. We find that the commonly used removal of $B_s$ by subtracting the toroidal average of magnetic perturbations intervenes in the Alfv\'en dynamics, causing spurious $E\times B$ transport. Instead, we propose an improved method to dynamically filter out the evolving background from the turbulent magnetic fluctuations in the time domain. The filter is verified in both low and high confinement tokamak conditions, confirming its capability to preserve the turbulence fidelity, provided sufficient filter width.

Accurate uncertainty information associated with essential climate variables (ECVs) is crucial for reliable climate modeling and understanding the spatiotemporal evolution of the Earth system. In recent years, geoscience and climate scientists have benefited from rapid progress in deep learning to advance the estimation of ECV products with improved accuracy. However, the quantification of uncertainties associated with the output of such deep learning models has yet to be thoroughly adopted. This survey explores the types of uncertainties associated with ECVs estimated from deep learning and the techniques to quantify them. The focus is on highlighting the importance of quantifying uncertainties inherent in ECV estimates, considering the dynamic and multifaceted nature of climate data. The survey starts by clarifying the definition of aleatoric and epistemic uncertainties and their roles in a typical satellite observation processing workflow, followed by bridging the gap between conventional statistical and deep learning views on uncertainties. Then, we comprehensively review the existing techniques for quantifying uncertainties associated with deep learning algorithms, focusing on their application in ECV studies. The specific need for modification to fit the requirements from both the Earth observation side and the deep learning side in such interdisciplinary tasks is discussed. Finally, we demonstrate our findings with two ECV examples, snow cover and terrestrial water storage, and provide our perspectives for future research.

This summary of the second Terrestrial Very-Long-Baseline Atom Interferometry (TVLBAI) Workshop provides a comprehensive overview of our meeting held in London in April 2024, building on the initial discussions during the inaugural workshop held at CERN in March 2023. Like the summary of the first workshop, this document records a critical milestone for the international atom interferometry community. It documents our concerted efforts to evaluate progress, address emerging challenges, and refine strategic directions for future large-scale atom interferometry projects. Our commitment to collaboration is manifested by the integration of diverse expertise and the coordination of international resources, all aimed at advancing the frontiers of atom interferometry physics and technology, as set out in a Memorandum of Understanding signed by over 50 institutions., Comment: Summary of the second Terrestrial Very-Long-Baseline Atom Interferometry Workshop held at Imperial College London: https://indico.cern.ch/event/1369392/

Depth completion upgrades sparse depth measurements into dense depth maps guided by a conventional image. Existing methods for this highly ill-posed task operate in tightly constrained settings and tend to struggle when applied to images outside the training domain or when the available depth measurements are sparse, irregularly distributed, or of varying density. Inspired by recent advances in monocular depth estimation, we reframe depth completion as an image-conditional depth map generation guided by sparse measurements. Our method, Marigold-DC, builds on a pretrained latent diffusion model for monocular depth estimation and injects the depth observations as test-time guidance via an optimization scheme that runs in tandem with the iterative inference of denoising diffusion. The method exhibits excellent zero-shot generalization across a diverse range of environments and handles even extremely sparse guidance effectively. Our results suggest that contemporary monocular depth priors greatly robustify depth completion: it may be better to view the task as recovering dense depth from (dense) image pixels, guided by sparse depth; rather than as inpainting (sparse) depth, guided by an image. Project website: https://MarigoldDepthCompletion.github.io/

We introduce an efficient open-source python package for the inverse design of three-dimensional photonic nanostructures using the Finite-Difference Time-Domain (FDTD) method. Leveraging a flexible reverse-mode automatic differentiation implementation, our software enables gradient-based optimization over large simulation volumes. Gradient computation is implemented within the JAX framework and based on the property of time reversibility in Maxwell's equations. This approach significantly reduces computational time and memory requirements compared to traditional FDTD methods. Gradient-based optimization facilitates the automatic creation of intricate three-dimensional structures with millions of design parameters, which would be infeasible to design manually. We demonstrate the scalability of the solver from single to multiple GPUs through several inverse design examples, highlighting its robustness and performance in large-scale photonic simulations. In addition, the package features an object-oriented and user-friendly API that simplifies the specification of materials, sources, and constraints. Specifically, it allows for intuitive positioning and sizing of objects in absolute or relative coordinates within the simulation scene. By rapid specification of the desired design properties and rapid optimization within the given user constraints, this open-source framework aims to accelerate innovation in photonic inverse design. It yields a powerful and accessible computational tool for researchers, applicable in a wide range of use cases, including but not limited to photonic waveguides, active devices, and photonic integrated circuits.

The model of localized fermions on the triangular lattice is analyzed in means of the Monte Carlo simulations in the grand canonical ensemble. The Hamiltonian of the system has a form of the extended Hubbard model (at the atomic limit) with nearest-neighbor Ising-like magnetic $J$ interactions and onsite Coulomb $U$ interactions. The model is investigated for both signs of $J$, arbitrary $U$ interaction and arbitrary chemical potential $\mu$ (or, equivalently, arbitrary particle concentration $n$). Based on the specific heat capacity and sublattice magnetization analyses, the phase diagrams of the model are determined. For ferromagnetic case ($J<0$), the transition from the ordered phase (which is a standard ferromagnet and can be stable up to $k_{B}T/|J| \approx 0.61$) is found to be second-order (for sufficiently large temperatures $k_{B}T/|J| \gtrsim 0.2$) or first-order (for $-10$, the ordered phase occurs in a range of $-1/2-1/2$ is always second-order for any model parameters. The ordered phase for $J>0$ can be stable up to $k_{B}T/|J| \approx 0.06$., Comment: 11 pages, 6 figures (8 panels), 71 references; RevTeX class, double-column formatting