Datasets:

librarian-bots

/

arxiv-metadata-snapshot

like 5

Follow

Librarian Bots 75

2501.04774

Kuldeep Deka

Nicol\'as Bernal, Kuldeep Deka, Marta Losada

Dark Matter Ultraviolet Freeze-in in General Reheating Scenarios

21 pages, 6 figures

hep-ph astro-ph.CO hep-th

http://creativecommons.org/licenses/by/4.0/

The dynamics of cosmic reheating, that is, on how the energy stored in the inflaton is transferred to the standard model (SM) thermal bath, is largely unknown. In this work, we show that the phenomenology of the nonbaryonic dark matter (DM) ultraviolet freeze-in production strongly depends on the dynamics of the cosmic-reheating era. Using a general parametrization for the Hubble expansion rate and SM temperature, we thoroughly investigate DM production during reheating, not only recovering earlier findings that focused on specific cases, but also exploring alternative scenarios. Additionally, we derive a generalized framework for DM production via inflaton decays and identify the viable parameter space, while simultaneously addressing constraints from CMB observations. As illustrative examples, we explore gravitational DM production through scatterings of SM particles or inflatons, deriving well-defined parameter regions for these scenarios.

2501.04775

Sushmita Agarwal

Sushmita Agarwa, Amit Shukla and Pranjali Sharma

Constraining $\gamma$-ray dissipation site in gravitationally lensed quasar -- PKS 1830$-$211

11 pages, 10 figures, 3 tables. Accepted for publication in MNRAS

astro-ph.HE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Variable $\gamma$-ray flares upto minute timescales reflect extreme particle acceleration sites. However, for high-redshift blazars, the detection of such rapid variations remains limited by current telescope sensitivities. Gravitationally lensed blazars serve as powerful tools to probe $\gamma$-ray production zones in distant sources, with time delays between lensed signals providing crucial insights into the spatial distribution of emission regions relative to the lens's mass-weighted center. We have utilized 15 years of Fermi-LAT $\gamma$-ray data from direction of PKS 1830$-$211 to understand the origin of flaring high-energy production zone at varying flux states. To efficiently estimate the (lensed) time delay, we used a machine learning-based tool - the Gaussian Process regression algorithm, in addition to - Autocorrelation function and Double power spectrum. We found a consistent time delay across all flaring activity states, indicating a similar location for the $\gamma$-ray emission zone, possibly within the radio core. The estimated time delay of approximately 20 days for the five flaring epochs was significantly shorter than previously estimated radio delays. This suggests that the $\gamma$-ray emission zone is closer to the central engine, in contrast to the radio emission zone, which is expected to be much farther away. A linear relationship between lag and magnification has been observed in the identified source and echo flares. Our results suggest that the $\gamma$-ray emission zone originates from similar regions away from the site of radio dissipation.

2501.04776

Erik Recio-Armengol

Erik Recio-Armengol, Joseph Bowles

IQPopt: Fast optimization of instantaneous quantum polynomial circuits in JAX

quant-ph

http://creativecommons.org/licenses/by/4.0/

IQPopt is a software package designed to optimize large-scale instantaneous quantum polynomial circuits on classical hardware. By exploiting an efficient classical simulation algorithm for expectation value estimation, circuits with thousands of qubits and millions of gates can be optimized, provided the relevant objective function has an efficient description in terms of Pauli-Z type observables. Since sampling from instantaneous quantum polynomial circuits is widely believed to be hard for classical computers, this provides a method to identify powerful circuit instances before deployment and sampling on quantum hardware, where computational advantages may exist. The package leverages automatic differentiation in JAX, can be accelerated with access to hardware accelerators such as graphics processing units, and contains a dedicated module that can be used to train and evaluate quantum generative models via the maximum mean discrepancy.

2501.04777

Kevin Nuckolls

Kevin P. Nuckolls, Michael G. Scheer, Dillon Wong, Myungchul Oh, Ryan L. Lee, Jonah Herzog-Arbeitman, Kenji Watanabe, Takashi Taniguchi, Biao Lian, Ali Yazdani

Spectroscopy of the Fractal Hofstadter Energy Spectrum

20 pages, 5 figures

cond-mat.mes-hall cond-mat.str-el

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Hofstadter's butterfly, the predicted energy spectrum for non-interacting electrons confined to a two-dimensional lattice in a magnetic field, is one of the most remarkable fractal structures in nature. At rational ratios of magnetic flux quanta per lattice unit cell, this spectrum shows self-similar distributions of energy levels that reflect its recursive construction. For most materials, Hofstadter's butterfly is predicted under experimental conditions that are unachievable using laboratory-scale magnetic fields. More recently, electrical transport studies have provided evidence for Hofstadter's butterfly in materials engineered to have artificially large lattice constants, such as those with moir\'e superlattices. Yet to-date, direct spectroscopy of the fractal energy spectrum predicted by Hofstadter nearly 50 years ago has remained out of reach. Here we use high-resolution scanning tunneling microscopy / spectroscopy (STM / STS) to probe the flat electronic bands in twisted bilayer graphene near the predicted second magic angle, an ideal setting for spectroscopic studies of Hofstadter's spectrum. Our study shows the fractionalization of flat moir\'e bands into discrete Hofstadter subbands and discerns experimental signatures of self-similarity of this spectrum. Moreover, our measurements uncover a spectrum that evolves dynamically with electron density, displaying phenomena beyond that of Hofstadter's original model due to the combined effects of strong correlations, Coulomb interactions, and the quantum degeneracy of electrons in twisted bilayer graphene.

2501.04778

Marco Dall'Amico

Marco Dall'Amico, Michela Mapelli, Giuliano Iorio, Guglielmo Costa, St\'ephane Charlot, Erika Korb, Cecilia Sgalletta, Marie Lecroq

Impact of accretion-induced chemically homogeneous evolution on stellar and compact binary populations

A&A submitted, comments welcome

astro-ph.HE astro-ph.GA astro-ph.SR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In binary star systems, mass transfer can spin up the accretor, possibly leading to efficient chemical mixing and chemically quasi-homogeneous evolution (CHE). Here, we explore the effects of accretion-induced CHE on both stellar populations and their compact binary remnants with the state-of-the-art population synthesis code SEVN. We find that CHE efficiently enhances the formation of Wolf-Rayet stars (WRs) from secondary stars, which are spun-up by accretion, while simultaneously preventing their evolution into red supergiant stars (RSGs). Including CHE in our models increases the fraction of WRs in our stellar sample by nearly a factor of $\approx3$ at low metallicity ($Z=0.001$). WRs formed through CHE are, on average, more massive and luminous than those formed without CHE. Most WRs formed via CHE end their life as black holes. As a direct consequence, the CHE mechanism enhances the formation of binary black holes (BBHs) and black hole-neutron star (BHNS) systems, while simultaneously quenching the production of binary neutron stars (BNSs). However, CHE significantly quenches the merger rate of BBHs, BHNSs and BNSs at low metallicity ($Z\leq{}0.004$), because most binary compact objects formed via CHE have large orbital periods. For instance, the number of BBH and BHNS mergers decreases by one order of magnitude at $Z=0.004$ in the CHE model compared to the standard scenario. Finally, we find that secondary stars experiencing CHE frequently produce the most massive compact object in the binary system. In BHNSs, this implies that the black hole progenitor is the secondary star. Conversely, BBHs formed through accretion-induced CHE likely have asymmetric black hole components, but only a negligible fraction of these asymmetric systems ultimately merge within an Hubble time.

2501.04779

Ana Bokuli\'c

Ana Bokuli\'c, Carlos A. R. Herdeiro

Exact multi-black hole spacetimes in Einstein-ModMax theory

9 pages

gr-qc

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Exact solutions describing multiple, electrically charged black holes (BHs) in a model of non-linear electrodynamics (NLE) minimally coupled to Einstein's gravity are presented. The NLE model is ModMax theory, that has attracted much attention due to its duality and conformal invariance, features shared with standard (linear) electrodynamics. In the non-extremal case, the solution has conical singularities, similarly to the multi Reissner-Nordstr\"om solution in Einstein-Maxwell theory. In the extremal case the solution is regular on and outside the event horizon; it is isometric to the Majumdar-Papapetrou solution, albeit the individual BHs have a non-unitary charge to mass ratio, due to screening effects. Using the ModMax electromagnetic duality invariance, magnetically charged and dyonic generalizations are also obtained. Finally, we construct multi-BH solutions with a positive cosmological constant.

2501.04780

Shubhayan Sarkar

Shubhayan Sarkar

Using quantum nonlocality for device-independent confirmation of relativistic effects

6 pages!! Comments are welcome:)

quant-ph

http://creativecommons.org/licenses/by/4.0/

Synchronizing clocks to measure time is a fundamental process underpinning every practical communication task from GPS to parallel computation. However, as the current protocols are based on classical communication between the sender and receiver, they are prone to simple attacks that could cause a slight delay in the signal which would then cause a massive error in further operations. In this work, we first explain a simple attack that in principle can cause an arbitrary delay in the signal between sender and receiver. We then propose a way to overcome this problem by using a recently contrived idea of device-independent certification which utilises quantum nonlocality. Consequently, clocks can be synchronized in a highly secure way without trusting any devices in the setup. We then extend this proposal to observe relativistic time dilation in a device-independent manner.

2501.04781

Emilio Vilches

Juan Guillermo Garrido, Maximiliano Lioi and Emilio Vilches

Inexact Catching-Up Algorithm for Moreau's Sweeping Processes

arXiv admin note: substantial text overlap with arXiv:2308.08093

math.OC cs.NA math.NA

http://creativecommons.org/licenses/by/4.0/

In this paper, we develop an inexact version of the catching-up algorithm for sweeping processes. We define a new notion of approximate projection, which is compatible with any numerical method for approximating exact projections, as this new notion is not restricted to remain strictly within the set. We provide several properties of the new approximate projections, which enable us to prove the convergence of the inexact catching-up algorithm in three general frameworks: prox-regular moving sets, subsmooth moving sets, and merely closed sets. Additionally, we apply our numerical results to address complementarity dynamical systems, particularly electrical circuits with ideal diodes. In this context, we implement the inexact catching-up algorithm using a primal-dual optimization method, which typically does not necessarily guarantee a feasible point. Our results are illustrated through an electrical circuit with ideal diodes. Our results recover classical existence results in the literature and provide new insights into the numerical simulation of sweeping processes.

2501.04782

Andrew Bond

Andrew Bond, Jui-Hsien Wang, Long Mai, Erkut Erdem, Aykut Erdem

GaussianVideo: Efficient Video Representation via Hierarchical Gaussian Splatting

10 pages, 10 figures

cs.CV

http://creativecommons.org/licenses/by-nc-sa/4.0/

Efficient neural representations for dynamic video scenes are critical for applications ranging from video compression to interactive simulations. Yet, existing methods often face challenges related to high memory usage, lengthy training times, and temporal consistency. To address these issues, we introduce a novel neural video representation that combines 3D Gaussian splatting with continuous camera motion modeling. By leveraging Neural ODEs, our approach learns smooth camera trajectories while maintaining an explicit 3D scene representation through Gaussians. Additionally, we introduce a spatiotemporal hierarchical learning strategy, progressively refining spatial and temporal features to enhance reconstruction quality and accelerate convergence. This memory-efficient approach achieves high-quality rendering at impressive speeds. Experimental results show that our hierarchical learning, combined with robust camera motion modeling, captures complex dynamic scenes with strong temporal consistency, achieving state-of-the-art performance across diverse video datasets in both high- and low-motion scenarios.

2501.04783

Chao Zhang

Chao Zhang, Yechen Li, Neha Arora, Damien Pierce, and Carolina Osorio

Traffic Simulations: Multi-City Calibration of Metropolitan Highway Networks

Published on the 27th IEEE International Conference on Intelligent Transportation Systems (ITSC) (2024)

cs.ET cs.SY eess.SY

http://creativecommons.org/licenses/by/4.0/

This paper proposes an approach to perform travel demand calibration for high-resolution stochastic traffic simulators. It employs abundant travel times at the path-level, departing from the standard practice of resorting to scarce segment-level sensor counts. The proposed approach is shown to tackle high-dimensional instances in a sample-efficient way. For the first time, case studies on 6 metropolitan highway networks are carried out, considering a total of 54 calibration scenarios. This is the first work to show the ability of a calibration algorithm to systematically scale across networks. Compared to the state-of-the-art simultaneous perturbation stochastic approximation (SPSA) algorithm, the proposed approach enhances fit to field data by an average 43.5% with a maximum improvement of 80.0%, and does so within fewer simulation calls.

2501.04784

Srikar Yellapragada

Srikar Yellapragada, Kowshik Thopalli, Vivek Narayanaswamy, Wesam Sakla, Yang Liu, Yamen Mubarka, Dimitris Samaras, Jayaraman J. Thiagarajan

Leveraging Registers in Vision Transformers for Robust Adaptation

Accepted at ICASSP 2025

cs.CV cs.LG

http://creativecommons.org/licenses/by/4.0/

Vision Transformers (ViTs) have shown success across a variety of tasks due to their ability to capture global image representations. Recent studies have identified the existence of high-norm tokens in ViTs, which can interfere with unsupervised object discovery. To address this, the use of "registers" which are additional tokens that isolate high norm patch tokens while capturing global image-level information has been proposed. While registers have been studied extensively for object discovery, their generalization properties particularly in out-of-distribution (OOD) scenarios, remains underexplored. In this paper, we examine the utility of register token embeddings in providing additional features for improving generalization and anomaly rejection. To that end, we propose a simple method that combines the special CLS token embedding commonly employed in ViTs with the average-pooled register embeddings to create feature representations which are subsequently used for training a downstream classifier. We find that this enhances OOD generalization and anomaly rejection, while maintaining in-distribution (ID) performance. Extensive experiments across multiple ViT backbones trained with and without registers reveal consistent improvements of 2-4\% in top-1 OOD accuracy and a 2-3\% reduction in false positive rates for anomaly detection. Importantly, these gains are achieved without additional computational overhead.

2501.04785

Aiden Sheckler

Aiden Sheckler

Anomaly Matching in 6d $\mathcal{N}=(2,0)$ SCFTs from M5 Cobordism

34 pp

hep-th

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We investigate the anomalies of 6d $\mathcal{N} = (2, 0)$ superconformal field theories for any ADE gauge group using the modern characterization of anomalies by cobordism. We propose that, in order to account for all features of the anomaly, a bordism theory with exotic tangential structure is needed. We then attempt to match the anomaly in the IR effective theory on the tensor branch with a suitable WZW term. Once again, we show that the exotic bordism structure is necessary to achieve this. Our results suggest a natural explanation for the origin of the Hopf-Wess-Zumino term.

2501.04786

Aabhas Gulati

Aabhas Gulati, Ion Nechita and Satvik Singh

Entanglement in cyclic sign invariant quantum states

quant-ph math-ph math.MP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We introduce and study bipartite quantum states that are invariant under the local action of the cyclic sign group. Due to symmetry, these states are sparse and can be parameterized by a triple of vectors. Their important semi-definite properties, such as positivity and positivity under partial transpose (PPT), can be simply characterized in terms of these vectors and their discrete Fourier transforms. We study in detail the entanglement properties of this family of symmetric states, showing in particular that it contains PPT entangled states. For states that are diagonal in the Dicke basis, deciding separability is equivalent to a circulant version of the complete positivity problem. We provide some geometric results for the PPT cone, showing in particular that it is polyhedral. In local dimension less than 5, we completely characterize these sets and construct entanglement witnesses; some partial results are also obtained for d = 6, 7. Finally, we initiate the study of cyclic sign covariant quantum channels, showing in particular that the PPT squared conjecture holds for some of these maps.

2501.04787

Frederic Jaron

F. Jaron, L. Baldreich, J. B\"ohm, P. Charlot, A. Collioud, J. Gruber, H. Kr\'asn\'a, I. Mart\'i-Vidal, A. Nothnagel, V. P\'erez-D\'iez

Mitigating Source Structure in Geodetic VLBI on the Visibility Level

4 pages, 6 figures, published in Proceedings of the 16th EVN Symposium, Ed. E. Ros, P. Benke, S.A. Dzib, I. Rottmann, & J.A. Zensus, Bonn: Max-Planck-Institut f\"ur Radioastronomie, 2024, pages 145-148

astro-ph.IM

http://creativecommons.org/licenses/by/4.0/

Geodetic and astrometric VLBI has entered a new era with the implementation of the VLBI Global Observing System (VGOS). These broadband and dual linear polarization observations aim at an accuracy of station coordinates of 1 mm and a reference frame stability of 0.1 mm/year. Although the extended brightness distribution of many of the radio-loud active galactic nuclei observed during geodetic VLBI sessions is resolved by the interferometer, the established processing chain still treats these objects as point sources. We investigate the impact of source structure on the visibility level and develop tools to remove the structure from the visibility data, right after correlation. Here we present our approach and show results obtained from observational VGOS data.

2501.04788

Maike Clausen

Maike Clausen, Ivelina Momcheva, Katherine E. Whitaker, Sam E. Cutler, Rachel S. Bezanson, James S. Dunlop, Norman A. Grogin, Anton M. Koekemoer, Derek McLeod, Ross McLure, Tim B. Miller, Erica Nelson, Arjen van der Wel, David Wake, Stijn Wuyts

The Evolution of Half-Mass Radii and Color Gradients for Young and Old Quiescent Galaxies at $0.5 < z < 3$ with JWST/PRIMER

15 pages, 9 Figures

astro-ph.GA astro-ph.CO

http://creativecommons.org/licenses/by/4.0/

We present a study of the size growth of the red sequence between $0.5<z<3,$ tracing the evolution of quiescent galaxies in both effective half-light and half-mass radii using multi-wavelength JWST/NIRCam imaging provided by the PRIMER survey. Half-light radii are measured from imaging in 6 different filters for 455 quiescent galaxies with log($M_*/M_{\odot}$)$>10$, whereas half-mass radii are derived from the F444W profiles together with the F277W-F444W color-$M_*$/L relation. We investigate the dependence of the ratio $r_{e, \mathrm{mass}}/r_{e, \mathrm{light}}$ on redshift, stellar mass, and the wavelength used to measure $r_{e, \mathrm{light}}$, also separating the sample into younger and older quiescent galaxies. Our data demonstrate that rest-frame infrared sizes accurately trace mass-weighted sizes while sizes measured at rest-frame optical wavelengths (0.5-0.7$\mu$m) are 0.1-0.2 dex larger, with only minor variations in redshift. We find that the average size of young quiescent galaxies agrees with that of old quiescent galaxies at intermediate masses, $10<$log($M_*/M_{\odot}$)$<11$, within their respective uncertainties in all observed-frame half-light, rest-frame half-light and half-mass radius measurements. At face value, our results point to a combination of progenitor bias and minor mergers driving the size growth of intermediate-mass quiescent galaxies at $0.5<z<3$. Our results further indicate that the varying contributions to the general quiescent population by young and old quiescent galaxies can mimic evolution in redshift.

2501.04789

Takeshi Ikuma

Takeshi Ikuma, Melda Kunduk, and Andrew J. McWhorter

Comparison of fundamental frequency estimators with subharmonic voice signals

9 pages, 6 figures

eess.AS cs.SD

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In clinical voice signal analysis, mishandling of subharmonic voicing may cause an acoustic parameter to signal false negatives. As such, the ability of a fundamental frequency estimator to identify speaking fundamental frequency is critical. This paper presents a sustained-vowel study, which used a quality-of-estimate classification to identify subharmonic errors and subharmonics-to-harmonics ratio (SHR) to measure the strength of subharmonic voicing. Five estimators were studied with a sustained vowel dataset: Praat, YAAPT, Harvest, CREPE, and FCN-F0. FCN-F0, a deep-learning model, performed the best both in overall accuracy and in correctly resolving subharmonic signals. CREPE and Harvest are also highly capable estimators for sustained vowel analysis.

2501.04790

Marco Cappa

Marco Cappa and Francesco Sciortino and Lorenzo Rovigatti

A phase-field model for DNA self-assembly

10 pages, 5 figures

cond-mat.soft

http://creativecommons.org/licenses/by/4.0/

We present a phase-field model based on the Cahn-Hilliard equation to investigate the kinetics of phase separation in DNA nanostar systems. Leveraging a realistic free-energy functional derived from Wertheim theory, our model captures the thermodynamic and dynamic properties of self-assembling DNA nanostars under various conditions. This approach allows for the study of both one-component and multi-component systems, including mixtures of different nanostar species and cross-linkers. Through numerical simulations, we demonstrate the model ability to replicate experimental observations, including liquid-liquid phase separation, surface tension variation, and the structural organization of multi-component systems. Our results highlight the versatility and predictive power of the Cahn-Hilliard framework, particularly for complex systems where detailed simulations are computationally prohibitive. This work provides a robust foundation for studying DNA-based materials and their potential applications in nanotechnology and biophysics, including liquid-liquid phase separation in cellular environments.

2501.04791

Adrian Schlosser

Tobias R\"ospel, Adrian Schlosser, Bj\"orn Malte Sch\"afer

Approximating non-Gaussian Bayesian partitions with normalising flows: statistics, inference and application to cosmology

astro-ph.CO astro-ph.IM

http://creativecommons.org/licenses/by/4.0/

Subject of this paper is the simplification of Markov chain Monte Carlo sampling as used in Bayesian statistical inference by means of normalising flows, a machine learning method which is able to construct an invertible and differentiable transformation between Gaussian and non-Gaussian random distributions. We use normalising flows to compute Bayesian partition functions for non-Gaussian distributions and show how normalising flows can be employed in finding analytical expressions for posterior distributions beyond the Gaussian limit. Flows offer advantages for the numerical evaluation of the partition function itself, as well as for cumulants and for the information entropy. We demonstrate how normalising flows in conjunction with Bayes partitions can be used in inference problems in cosmology and apply them to the posterior distribution for the matter density $\Omega_m$ and a dark energy equation of state parameter $w_0$ on the basis of supernova data.

2501.04792

Zeinab Askari Donbeh

Zeinab Askari Donbeh (1), Mehdi Rasti (1), Shiva Kazemi Taskooh (1), and Mehdi Monemi (1) ((1) Centre for Wireless Communications (CWC), University of Oulu, Oulu, Finland)

An Analysis on Stabilizability and Reliability Relationship in Wireless Networked Control Systems

Accepted for WCNC-2025 conference

cs.NI

http://creativecommons.org/licenses/by/4.0/

The stabilizability of wireless networked control systems (WNCSs) is a deterministic binary valued parameter proven to hold if the communication data rate is higher than the sum of the logarithm of unstable eigenvalues of the open-loop control system. In this analysis, it is assumed that the communication system provides a fixed deterministic transmission rate between the sensors and controllers. Due to the stochastic parameters of communication channels, such as small-scale fading, the instantaneous rate is an intrinsically stochastic parameter. In this sense, it is a common practice in the literature to use the deterministic ergodic rate in analyzing the asymptotic stabilizability. Theoretically, there exists no work in the literature investigating how the ergodic rate can be incorporated into the analysis of asymptotic stabilizability. Considering the stochastic nature of channel parameters, we introduce the concept of probability of stabilizability by interconnecting communication link reliability with the system's unstable eigenvalues and derive a closed-form expression that quantifies this metric. Numerical results are provided to visualize how communication and control systems' parameters affect the probability of stabilizability of the overall system.

2501.04793

Caner Odabas

Caner Odaba\c{s} and \"Omer Morg\"ul

A Novel Observer Design for LuGre Friction Estimation and Control

eess.SY cs.SY

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Dynamic components of the friction may directly impact the stability and performance of the motion control systems. The LuGre model is a prevalent friction model utilized to express this dynamic behavior. Since the LuGre model is very comprehensive, friction compensation based on it might be challenging. Inspired by this, we develop a novel observer to estimate and compensate for LuGre friction. Furthermore, we present a Lyapunov stability analysis to show that observer dynamics are asymptotically stable under certain conditions. Compared to its counterparts, the proposed observer constitutes a simple and standalone scheme that can be utilized with arbitrary control inputs in a straightforward way. As a primary difference, the presented observer estimates velocity and uses the velocity error to estimate friction in addition to control input. The extensive simulations revealed that the introduced observer enhances position and velocity tracking performance in the presence of friction.

2501.04794

Gianfranco Cort\'es

Gianfranco Cortes and Baba C. Vemuri

A Steerable Deep Network for Model-Free Diffusion MRI Registration

eess.IV cs.CV cs.LG

http://creativecommons.org/licenses/by/4.0/

Nonrigid registration is vital to medical image analysis but remains challenging for diffusion MRI (dMRI) due to its high-dimensional, orientation-dependent nature. While classical methods are accurate, they are computationally demanding, and deep neural networks, though efficient, have been underexplored for nonrigid dMRI registration compared to structural imaging. We present a novel, deep learning framework for model-free, nonrigid registration of raw diffusion MRI data that does not require explicit reorientation. Unlike previous methods relying on derived representations such as diffusion tensors or fiber orientation distribution functions, in our approach, we formulate the registration as an equivariant diffeomorphism of position-and-orientation space. Central to our method is an $\mathsf{SE}(3)$-equivariant UNet that generates velocity fields while preserving the geometric properties of a raw dMRI's domain. We introduce a new loss function based on the maximum mean discrepancy in Fourier space, implicitly matching ensemble average propagators across images. Experimental results on Human Connectome Project dMRI data demonstrate competitive performance compared to state-of-the-art approaches, with the added advantage of bypassing the overhead for estimating derived representations. This work establishes a foundation for data-driven, geometry-aware dMRI registration directly in the acquisition space.

2501.04795

Xueda Wen

Jiyuan Fang, Qi Zhou, Xueda Wen

Phase Transitions in Quasi-Periodically Driven Quantum Critical Systems: Analytical Results

20 pages, many figures

cond-mat.stat-mech cond-mat.str-el hep-th math-ph math.MP quant-ph

http://creativecommons.org/licenses/by/4.0/

In this work, we study analytically the phase transitions in quasi-periodically driven one dimensional quantum critical systems that are described by conformal field theories (CFTs). The phase diagrams and phase transitions can be analytically obtained by using Avila's global theory in one-frequency quasiperiodic cocycles. Compared to the previous works where the quasiperiodicity was introduced in the driving time and no phase transitions were observed [1], here we propose a setup where the quasiperiodicity is introduced in the driving Hamiltonians. In our setup, one can observe the heating phases, non-heating phases, and the phase transitions. The phase diagram as well as the Lyapunov exponents that determine the entanglement entropy evolution can be analytically obtained. In addition, based on Avila's theory, we prove there is no phase transition in the previously proposed setup of quasi-periodically driven CFTs [1]. We verify our field theory results by studying the time evolution of entanglement entropy on lattice models.

2501.04796

M. Amin Rahimian

M. Amin Rahimian and Michael P. Colaresi

Democratic Resilience and Sociotechnical Shocks

Computational and Mathematical Organization Theory, forthcoming

cs.SI cs.SY eess.SY stat.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We focus on the potential fragility of democratic elections given modern information-communication technologies (ICT) in the Web 2.0 era. Our work provides an explanation for the cascading attrition of public officials recently in the United States and offers potential policy interventions from a dynamic system's perspective. We propose that micro-level heterogeneity across individuals within crucial institutions leads to vulnerabilities of election support systems at the macro scale. Our analysis provides comparative statistics to measure the fragility of systems against targeted harassment, disinformation campaigns, and other adversarial manipulations that are now cheaper to scale and deploy. Our analysis also informs policy interventions that seek to retain public officials and increase voter turnout. We show how limited resources (for example, salary incentives to public officials and targeted interventions to increase voter turnout) can be allocated at the population level to improve these outcomes and maximally enhance democratic resilience. On the one hand, structural and individual heterogeneity cause systemic fragility that adversarial actors can exploit, but also provide opportunities for effective interventions that offer significant global improvements from limited and localized actions.

2501.04797

Karlo Lelas

Karlo Lelas and Nikola Poljak

Driven damped harmonic oscillator revisited: energy resonance

are welcome

physics.class-ph

http://creativecommons.org/licenses/by/4.0/

We derive the exact expression for the resonant frequency of the time-averaged steady-state energy and show that this frequency is excellently approximated by the arithmetic mean of the amplitude and velocity resonant frequencies. In addition, we argue that the frequency of the amplitude resonance can be regarded as the energy resonance frequency, since it provides the maximal peak values of instantaneous energy.

2501.04798

Breno Bernard Nicolau De Fran\c{c}a

Breno Bernard Nicolau de Fran\c{c}a, Dietmar Pfahl, Valdemar Vicente Graciano Neto, Nauman bin Ali

Teaching Simulation as a Research Method in Empirical Software Engineering

cs.SE

http://creativecommons.org/licenses/by-nc-nd/4.0/

The chapter supports educators and postgraduate students in understanding the role of simulation in software engineering research based on the authors' experience. This way, it includes a background positioning simulation-based studies in software engineering research, the proposition of learning objectives for teaching simulation as a research method, and presents our experience when teaching simulation concepts and practice. For educators, it further provides learning objectives when teaching simulation, considering the current state of the art in software engineering research and the necessary guidance and recommended learning activities to achieve these objectives. For students, it drives the learning path for those interested in learning this method but had no opportunity to engage in an entire course on simulation in the context of empirical research.

2501.04799

Sanjana Sankar

Sanjana Sankar, Martin Lenglet, Gerard Bailly, Denis Beautemps, Thomas Hueber

Cued Speech Generation Leveraging a Pre-trained Audiovisual Text-to-Speech Model

ICASSP 2025 - IEEE International Conference on Acoustics, Speech and Signal Processing, April 2025, Hyderabad, India

cs.CL

http://creativecommons.org/licenses/by/4.0/

This paper presents a novel approach for the automatic generation of Cued Speech (ACSG), a visual communication system used by people with hearing impairment to better elicit the spoken language. We explore transfer learning strategies by leveraging a pre-trained audiovisual autoregressive text-to-speech model (AVTacotron2). This model is reprogrammed to infer Cued Speech (CS) hand and lip movements from text input. Experiments are conducted on two publicly available datasets, including one recorded specifically for this study. Performance is assessed using an automatic CS recognition system. With a decoding accuracy at the phonetic level reaching approximately 77%, the results demonstrate the effectiveness of our approach.

2501.04800

Abigail Moran

Abigail Moran, Lorenzo Sironi, Aviad Levis, Bart Ripperda, Elias R. Most, Sebastiaan Selvi

Effective resistivity in relativistic reconnection: a prescription based on fully kinetic simulations

astro-ph.HE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

A variety of high-energy astrophysical phenomena are powered by the release -- via magnetic reconnection -- of the energy stored in oppositely directed fields. Single-fluid resistive magnetohydrodynamic (MHD) simulations with uniform resistivity yield dissipation rates that are much lower (by nearly one order of magnitude) than equivalent kinetic calculations. Reconnection-driven phenomena could be accordingly modeled in resistive MHD employing a non-uniform, ``effective'' resistivity informed by kinetic calculations. In this work, we analyze a suite of fully kinetic particle-in-cell (PIC) simulations of relativistic pair-plasma reconnection -- where the magnetic energy is greater than the rest mass energy -- for different strengths of the guide field orthogonal to the alternating component. We extract an empirical prescription for the effective resistivity, $\eta_{\mathrm{eff}} = \alpha B_0 \mathbf{|J|}^p / \left(|\mathbf{J}|^{p+1}+\left(e n_t c\right)^{p+1}\right)$, where $B_0$ is the reconnecting magnetic field strength, $\bf J$ is the current density, $n_t$ the lab-frame total number density, $e$ the elementary charge, and $c$ the speed of light. The guide field dependence is encoded in $\alpha$ and $p$, which we fit to PIC data. This resistivity formulation -- which relies only on single-fluid MHD quantities -- successfully reproduces the spatial structure and strength of nonideal electric fields, and thus provides a promising strategy for enhancing the reconnection rate in resistive MHD simulations.

2501.04801

Matthew Low

Tao Han, Matthew Low, Youle Su

Entanglement and Bell Nonlocality in $\tau^+ \tau^-$ at the BEPC

28 pages, 7 figures, 5 tables

PITT-PACC-2412

hep-ph hep-ex

http://creativecommons.org/licenses/by/4.0/

Quantum entanglement and Bell nonlocality are two phenomena that occur only in quantum systems. In both cases, these are correlations between two subsystems that are classically absent. Traditionally, these phenomena have been measured in low-energy photon and electron experiments, but more recently they have also been measured in high-energy particle collider environments. In this work, we propose measuring the entanglement and Bell nonlocality in the $\tau^+\tau^-$ state near and above its kinematic threshold at the Beijing Electron Positron Collider (BEPC). We find that in the existing dataset, entanglement is observable if systematic uncertainties are kept to 1%. In the upcoming run between 4.0 and 5.6 GeV, the entanglement is predicted to be measurable with a precision better than 4% and Bell nonlocality can be established at $5\sigma$ as long as systematic uncertainty can be controlled at level of 0.5% - 2.0%, depending on the center-of-mass energy.

2501.04802

Yongkang Li

Yongkang Li, Panagiotis Eustratiadis, Evangelos Kanoulas

Reproducing HotFlip for Corpus Poisoning Attacks in Dense Retrieval

This paper has been accepted for oral presentation in the reproducibility track at ECIR 2025

cs.IR cs.CL

http://creativecommons.org/licenses/by/4.0/

HotFlip is a topical gradient-based word substitution method for attacking language models. Recently, this method has been further applied to attack retrieval systems by generating malicious passages that are injected into a corpus, i.e., corpus poisoning. However, HotFlip is known to be computationally inefficient, with the majority of time being spent on gradient accumulation for each query-passage pair during the adversarial token generation phase, making it impossible to generate an adequate number of adversarial passages in a reasonable amount of time. Moreover, the attack method itself assumes access to a set of user queries, a strong assumption that does not correspond to how real-world adversarial attacks are usually performed. In this paper, we first significantly boost the efficiency of HotFlip, reducing the adversarial generation process from 4 hours per document to only 15 minutes, using the same hardware. We further contribute experiments and analysis on two additional tasks: (1) transfer-based black-box attacks, and (2) query-agnostic attacks. Whenever possible, we provide comparisons between the original method and our improved version. Our experiments demonstrate that HotFlip can effectively attack a variety of dense retrievers, with an observed trend that its attack performance diminishes against more advanced and recent methods. Interestingly, we observe that while HotFlip performs poorly in a black-box setting, indicating limited capacity for generalization, in query-agnostic scenarios its performance is correlated to the volume of injected adversarial passages.

2501.04803

Nirvana Coppola

Emiliano Ambrosi, Nirvana Coppola, Francesc Fit\'e

Geometrically simple counterexamples to a local-global principle for quadratic twists

15 pages. Comments are welcome!

math.NT math.AG

http://creativecommons.org/licenses/by/4.0/

Two abelian varieties $A$ and $B$ over a number field $K$ are said to be strongly locally quadratic twists if they are quadratic twists at every completion of $K$. While it was known that this does not imply that $A$ and $B$ are quadratic twists over $K$, the only known counterexamples (necessarily of dimension $\geq 4$) are not geometrically simple. We show that, for every prime $p\equiv 13 \pmod{24}$, there exists a pair of geometrically simple abelian varieties of dimension $p-1$ over $\mathbb{Q}$ that are strongly locally quadratic twists but not quadratic twists. The proof is based on Galois cohomology computations and class field theory.

2501.04804

Hamed Khaleghi

Hamed Khaleghi, Didem Ozevin, Aslihan Karatas

Non-Destructive Argon Concentration Assessment within Insulating Glass Units (IGUs) using Ultrasonic Technique

physics.ins-det

http://creativecommons.org/licenses/by-sa/4.0/

Insulating glass units (IGUs) contribute significantly to energy loss among building envelopes, as they are responsible for approximately 30 to 50% of thermal transmission losses. To mitigate these losses, inert gases such as Argon are used within IGU spacers. The aging of IGUs cause decrease in Argon gas concentration, affecting their insulating properties. Therefore, it is important to evaluate the Argon concentration to maintain the insulating effectiveness of IGUs. Therefore, this study focuses on utilizing ultrasonic testing (UT) to develop a non-destructive and affordable methodology for assessing argon concentration within IGUs. The methodology is developed in four steps: (1) designing an experimental setup for creating accurate gas mixtures of Argon and air, and transferring the target gas mixtures to the IGU spacer; (2) performing ultrasonic tests on IGUs with 60kHz excitation frequency to generate UT waveforms; (3) performing a validation test of the proposed methodology to verify the accuracy and repeatability of results obtained from step 2; and (4) analyzing the ultrasonic waveforms to explore the correlation between UT properties and Argon concentration within IGU. The findings show that ultrasonic energy serves as the most effective UT feature for assessing Argon concentration within the IGU spacer, as evidenced by a strong correlation with an R-squared value exceeding 0.9 across all the UT measurements. This research introduces a user-friendly solution that enhances the ease of assessing Argon concentration within IGUs' spacers, thereby facilitating the monitoring of their operational performance. The proposed methodology for evaluating Argon concentration holds potential benefits for IGU manufacturers and researchers. Moreover, it serves as a valuable tool for improving quality control across the manufacturing, maintenance, and operational stages of IGUs.

2501.04805

Alberto Del Pia

Alberto Del Pia and Aida Khajavirad

Extended formulations for the multilinear polytope of acyclic hypergraphs

arXiv admin note: text overlap with arXiv:2212.11239

math.OC cs.DM

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This article provides an overview of our joint work on binary polynomial optimization over the past decade. We define the multilinear polytope as the convex hull of the feasible region of a linearized binary polynomial optimization problem. By representing the multilinear polytope with hypergraphs, we investigate the connections between hypergraph acyclicity and the complexity of the facial structure of the multilinear polytope. We characterize the acyclic hypergraphs for which a polynomial-size extended formulation for the multilinear polytope can be constructed in polynomial time.

2501.04806

Wei-Chun Jao

Wei-Chun Jao and Allison Youngblood

Mind the Gap II: the near-UV fluxes of M dwarfs

20 pages, 10 figures, 5 tables, accepted by AJ

astro-ph.SR

http://creativecommons.org/licenses/by/4.0/

Because of the continuous variations in mass, metallicity, and opacity, dwarf stars are distributed along the main sequence on optical and near-IR color-magnitude diagrams following a smooth polynomial. In this study of utilizing a catalog of cross-matched GALEX and Gaia sources, we identify two distinct populations of M dwarfs in the near-ultraviolet (NUV) band on the $M_{NUV}$ vs. $M_G$ diagram. We also reveal a pronounced increase in the number of stars exhibiting high NUV fluxes near the spectral type M2 or $M_G\sim9.4$, coinciding with the $H_2$ formation in the atmosphere to improve the energy transportation at the surface. This suggests that certain yet-to-be-understood stellar mechanisms drive heightened activity in the NUV band around the effective temperature of M2 and later types of M dwarfs. Through examination of archival Hubble Space Telescope spectra, we show that Fe II line forests at $\sim$2400A and 2800A dominate the spectral features in the GALEX NUV bandpass, contributing to the observed excess fluxes at a given mass between the two populations. Additionally, our investigation indicates that fast rotators and young stars likely increase brightness in the NUV band, but not all stars with bright NUV fluxes are fast rotators or young stars.

2501.04807

Andrei Tokovinin

Andrei Tokovinin

Orbits of Six Triple Systems

13 pages, 9 Figures. Accepted by The Astronomical Journal

astro-ph.SR

http://creativecommons.org/licenses/by/4.0/

Joint analysis of position measurements and radial velocities of six triple stellar systems is conducted to determine their inner and/or outer orbits. Accumulation of such data is needed to study the architecture of stellar hierarchies and its relation to the formation mechanisms. The inner periods in the six systems (HIP 11783, 64836, 72423, 84720, 89234, and 105404) range from 0.5 days to 44 yr. The shortest outer period of 3.34 yr is found in the compact triple HIP~105404 (BS Ind). The resolved triple system HIP 64836 has comparable inner and outer periods (5 and 30 yr), placing it near the limit of dynamical stability, while its quasi-circular and coplanar orbits suggest a 1:6 mean motion resonance. The periods in HIP 89234 (44 and ~450 yr) are also comparable, but the mutual orbit inclination is large, 54 degrees. Masses of the components are estimated and each system is discussed individually.

2501.04808

J\'er\'emy Sourd

J. Sourd, T. Kotte, P. Wild, S. M\"uhlbauer, J. Wosnitza, S. Zherlitsyn

Rotational magnetoelastic interactions in the Dzyaloshinskii-Moriya magnet Ba$_2$CuGe$_2$O$_7$

cond-mat.str-el

http://creativecommons.org/licenses/by/4.0/

We report the magnetoelastic properties of a Ba$_2$CuGe$_2$O$_7$ single crystal at low temperatures under a magnetic field applied along the crystallographic [001] axis. Our results extend to low temperature the $H-T$ phase diagram determined for this compound by neutron scattering. Furthermore, we observe that specific elastic modes are better sensitive to the various magnetic transitions. In particular, we observe an unusual coupling between the in-plane transverse acoustic mode and the cycloidal order at low field, which suggests a novel spin-strain mechanism originating from Dzyaloshinskii-Moriya interaction in this compound.

2501.04809

Joerg Schumacher

Shadab Alam, Dmitry Krasnov, Ambrish Pandey, John Panickacheril John, Roshan J. Samuel, Philipp P. Vieweg, J\"org Schumacher

Turbulent mesoscale convection in the Boussinesq limit and beyond

26 pages, 16 figures (some in reduced quality)

physics.flu-dyn

http://creativecommons.org/licenses/by/4.0/

Mesoscale convection covers an intermediate scale range between small-scale turbulence and the global organization of the convection flow. It is often characterized by an order of the convection patterns despite very high Rayleigh numbers and strong turbulent fluctuations. In this review, we discuss several aspects of mesoscale convection, which have been investigated by three-dimensional direct numerical simulations. The numerical studies are performed in a characteristic configuration of a plane layer that is heated from below and cooled from above or subject to constant heat flux at the top and bottom boundaries. We discuss the role of the thermal and mechanical boundary conditions for structure formation and study the impact of the domain shape as well as the Prandtl number. With respect to the latter, we focus on low values that arise in astrophysical convection and are partly not anymore accessible in laboratory experiments with liquid metals. Beside these experiments in the Boussinesq approximation, we report studies of non-Boussinesq mesoscale convection. This is done by investigating effects of compressibility and temperature dependence of material properties. The kinetic energy dissipation rate turns out to remain a central quantity for the turbulent mixing in compressible convection. Their different components, statistics, relation to the turbulent viscosity, and the multifractal properties are discussed.

2501.04810

Andreas Vogelsang

Andreas Vogelsang, Alexander Korn, Giovanna Broccia, Alessio Ferrari, Jannik Fischbach, Chetan Arora

On the Impact of Requirements Smells in Prompts: The Case of Automated Traceability

Accepted at 2025 IEEE/ACM 47th International Conference on Software Engineering: New Ideas and Emerging Results (ICSE-NIER)

cs.SE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Large language models (LLMs) are increasingly used to generate software artifacts, such as source code, tests, and trace links. Requirements play a central role in shaping the input prompts that guide LLMs, as they are often used as part of the prompts to synthesize the artifacts. However, the impact of requirements formulation on LLM performance remains unclear. In this paper, we investigate the role of requirements smells-indicators of potential issues like ambiguity and inconsistency-when used in prompts for LLMs. We conducted experiments using two LLMs focusing on automated trace link generation between requirements and code. Our results show mixed outcomes: while requirements smells had a small but significant effect when predicting whether a requirement was implemented in a piece of code (i.e., a trace link exists), no significant effect was observed when tracing the requirements with the associated lines of code. These findings suggest that requirements smells can affect LLM performance in certain SE tasks but may not uniformly impact all tasks. We highlight the need for further research to understand these nuances and propose future work toward developing guidelines for mitigating the negative effects of requirements smells in AI-driven SE processes.

2501.04811

Luke Dramko

Luke Dramko, Claire Le Goues, Edward J. Schwartz

Fast, Fine-Grained Equivalence Checking for Neural Decompilers

cs.LG cs.CR cs.SE

http://creativecommons.org/licenses/by-sa/4.0/

Neural decompilers are machine learning models that reconstruct the source code from an executable program. Critical to the lifecycle of any machine learning model is an evaluation of its effectiveness. However, existing techniques for evaluating neural decompilation models have substantial weaknesses, especially when it comes to showing the correctness of the neural decompiler's predictions. To address this, we introduce codealign, a novel instruction-level code equivalence technique designed for neural decompilers. We provide a formal definition of a relation between equivalent instructions, which we term an equivalence alignment. We show how codealign generates equivalence alignments, then evaluate codealign by comparing it with symbolic execution. Finally, we show how the information codealign provides-which parts of the functions are equivalent and how well the variable names match-is substantially more detailed than existing state-of-the-art evaluation metrics, which report unitless numbers measuring similarity.

2501.04812

Kalman Varga

Ankita Nair, Vikas Bharti, Yetmgeta S. Aklilu, and Kalman Varga

The effect of finite mass in cavity QED calculations

physics.chem-ph

http://creativecommons.org/licenses/by/4.0/

The effect of finite nuclear mass is investigated in coupled light matter systems in cavity quantum electrodynamics (cavity QED) using the Pauli-Fierz Hamiltonian. Three different systems, the He atom, the H$^-$ ion and the H$_2^+$ ion is investigated. There are small, but significant differences in the behavior of the binding energies as the function of the coupling strength. The probability of coupling to light is found to be very small but even this small coupling has a very strong effect on the energies of the systems.

2501.04813

Ermiya Farokhnejad

Sharareh Alipour, Ermiya Farokhnejad, Tobias M\"omke

Improved Approximation Algorithms for (1,2)-TSP and Max-TSP Using Path Covers in the Semi-Streaming Model

To appear in STACS 2025

cs.DS

http://creativecommons.org/licenses/by/4.0/

We investigate semi-streaming algorithms for the Traveling Salesman Problem (TSP). Specifically, we focus on a variant known as the $(1,2)$-TSP, where the distances between any two vertices are either one or two. Our primary emphasis is on the closely related Maximum Path Cover Problem, which aims to find a collection of vertex-disjoint paths that covers the maximum number of edges in a graph. We propose an algorithm that, for any $\epsilon > 0$, achieves a $(\frac{2}{3}-\epsilon)$-approximation of the maximum path cover size for an $n$-vertex graph, using $\text{poly}(\frac{1}{\epsilon})$ passes. This result improves upon the previous $\frac{1}{2}$-approximation by Behnezhad and et al. [ICALP 2024] in the semi-streaming model. Building on this result, we design a semi-streaming algorithm that constructs a tour for an instance of \tsp with an approximation factor of $(\frac{4}{3} + \epsilon)$, improving upon the previous $\frac{3}{2}$-approximation actor algorithm by Behnezhad and et al. [ICALP 2024] (Although it is not explicitly stated in this paper that their algorithm works in the semi-streaming model, it is easy to verify). Furthermore, we extend our approach to develop an approximation algorithm for the Maximum TSP (Max-TSP), where the goal is to find a Hamiltonian cycle with the maximum possible weight in a given weighted graph $G$. Our algorithm provides a $(\frac{7}{12} - \epsilon)$-approximation for Max-TSP in $\text{poly}(\frac{1}{\epsilon})$ passes, improving on the previously known $(\frac{1}{2}-\epsilon)$-approximation obtained via maximum weight matching in the semi-streaming model.

2501.04814

Brice Flamencourt

Brice Flamencourt, Abdelghani Zeghib

On foliations admitting a transverse similarity structure

21 pages

math.DG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We give a "conceptual" approach to Kourganoff's results about foliations with a transverse similarity structure. In particular, we give a proof, understandable by the targeted community, of the very important result classifying the holonomy of the closed, non-exact Weyl structures on compact manifolds, from which arose the notion of locally conformally product structures. We also extract from the proof several results on foliations admitting locally metric transverse connections.

2501.04815

Kaouther Messaoud

Kaouther Messaoud, Matthieu Cord, Alexandre Alahi

Towards Generalizable Trajectory Prediction Using Dual-Level Representation Learning And Adaptive Prompting

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Existing vehicle trajectory prediction models struggle with generalizability, prediction uncertainties, and handling complex interactions. It is often due to limitations like complex architectures customized for a specific dataset and inefficient multimodal handling. We propose Perceiver with Register queries (PerReg+), a novel trajectory prediction framework that introduces: (1) Dual-Level Representation Learning via Self-Distillation (SD) and Masked Reconstruction (MR), capturing global context and fine-grained details. Additionally, our approach of reconstructing segmentlevel trajectories and lane segments from masked inputs with query drop, enables effective use of contextual information and improves generalization; (2) Enhanced Multimodality using register-based queries and pretraining, eliminating the need for clustering and suppression; and (3) Adaptive Prompt Tuning during fine-tuning, freezing the main architecture and optimizing a small number of prompts for efficient adaptation. PerReg+ sets a new state-of-the-art performance on nuScenes [1], Argoverse 2 [2], and Waymo Open Motion Dataset (WOMD) [3]. Remarkable, our pretrained model reduces the error by 6.8% on smaller datasets, and multi-dataset training enhances generalization. In cross-domain tests, PerReg+ reduces B-FDE by 11.8% compared to its non-pretrained variant.

2501.04816

Felix Jimenez

Felix Jimenez, Matthias Katzfuss

Probabilistic Skip Connections for Deterministic Uncertainty Quantification in Deep Neural Networks

15 pages, 9 figures

cs.LG stat.ML

http://creativecommons.org/licenses/by/4.0/

Deterministic uncertainty quantification (UQ) in deep learning aims to estimate uncertainty with a single pass through a network by leveraging outputs from the network's feature extractor. Existing methods require that the feature extractor be both sensitive and smooth, ensuring meaningful input changes produce meaningful changes in feature vectors. Smoothness enables generalization, while sensitivity prevents feature collapse, where distinct inputs are mapped to identical feature vectors. To meet these requirements, current deterministic methods often retrain networks with spectral normalization. Instead of modifying training, we propose using measures of neural collapse to identify an existing intermediate layer that is both sensitive and smooth. We then fit a probabilistic model to the feature vector of this intermediate layer, which we call a probabilistic skip connection (PSC). Through empirical analysis, we explore the impact of spectral normalization on neural collapse and demonstrate that PSCs can effectively disentangle aleatoric and epistemic uncertainty. Additionally, we show that PSCs achieve uncertainty quantification and out-of-distribution (OOD) detection performance that matches or exceeds existing single-pass methods requiring training modifications. By retrofitting existing models, PSCs enable high-quality UQ and OOD capabilities without retraining.

2501.04817

Eiman Kanjo Prof.

Ziyuan Bao, Eiman Kanjo, Soumya Banerjee, Hasib-Al Rashid, Tinoosh Mohsenin

Decentralised Resource Sharing in TinyML: Wireless Bilayer Gossip Parallel SGD for Collaborative Learning

cs.LG cs.AI

http://creativecommons.org/licenses/by-nc-nd/4.0/

With the growing computational capabilities of microcontroller units (MCUs), edge devices can now support machine learning models. However, deploying decentralised federated learning (DFL) on such devices presents key challenges, including intermittent connectivity, limited communication range, and dynamic network topologies. This paper proposes a novel framework, bilayer Gossip Decentralised Parallel Stochastic Gradient Descent (GD PSGD), designed to address these issues in resource-constrained environments. The framework incorporates a hierarchical communication structure using Distributed Kmeans (DKmeans) clustering for geographic grouping and a gossip protocol for efficient model aggregation across two layers: intra-cluster and inter-cluster. We evaluate the framework's performance against the Centralised Federated Learning (CFL) baseline using the MCUNet model on the CIFAR-10 dataset under IID and Non-IID conditions. Results demonstrate that the proposed method achieves comparable accuracy to CFL on IID datasets, requiring only 1.8 additional rounds for convergence. On Non-IID datasets, the accuracy loss remains under 8\% for moderate data imbalance. These findings highlight the framework's potential to support scalable and privacy-preserving learning on edge devices with minimal performance trade-offs.

2501.04818

Charlie Lloyd

Charlie Lloyd and Robert Dorrell

Transport by waves and turbulence: Dilute suspensions in stably stratified plane Poiseuille flow

physics.flu-dyn

http://creativecommons.org/licenses/by/4.0/

We investigate the transport dynamics of negatively buoyant passive sediment in a thermally stratified plane Poiseuille flow using numerical simulation, exploring the influence of large-scale coherent structures on sediment transport processes. Sediment concentration fields are passively solved with settling velocities of $V_s = 0.005$, $V_s = 0.01$ and $V_s = 0.02$, made dimensionless by the friction velocity. Strong buoyancy forces in the core have a profound impact on sediment transport, leading to two-layer sediment concentration profiles with concentration gradients increasing with sediment settling velocity, $V_s$. When compared to unstratified flow, stratification leads to considerably larger differences in concentration profiles and higher order statistics between the three different sediments. When scalar statistics are appropriately scaled by either $\theta_\tau$ for temperature or $V_s \overline{c}$ for sediment, where $\theta_\tau$ is the thermal shear temperature and $\overline{c}$ is the vertically varying mean sediment concentration, scalar statistics collapse to near common vertical profiles. Collapse is explained by revisiting the gradient diffusion hypothesis, which links scalar fluxes to respective mean gradients. However, collapse of scalars is poor in the channel core where large concentration gradients coincide with large-scale mixing events. Here the differences between vertical turbulent diffusivities of sediment and temperature increase with increasing $V_s$, reaching 20% for $V_s = 0.02$. Discrepancies arise due to a breakdown of the linear gradient diffusion hypothesis. Predictions using the gradient diffusion hypothesis are expected to worsen with increasing settling velocity, indicating that classical models poorly predict dilute particle transport in strongly stratified flows, which are common in a range of environmental and industrial settings.

2501.04819

Anthony Desch\^enes

Anthony Desch\^enes, R\'emi Georges, Cem Subakan, Bruna Ugulino, Antoine Henry, Michael Morin

Planing It by Ear: Convolutional Neural Networks for Acoustic Anomaly Detection in Industrial Wood Planers

cs.SD cs.AI eess.AS

http://creativecommons.org/licenses/by/4.0/

In recent years, the wood product industry has been facing a skilled labor shortage. The result is more frequent sudden failures, resulting in additional costs for these companies already operating in a very competitive market. Moreover, sawmills are challenging environments for machinery and sensors. Given that experienced machine operators may be able to diagnose defects or malfunctions, one possible way of assisting novice operators is through acoustic monitoring. As a step towards the automation of wood-processing equipment and decision support systems for machine operators, in this paper, we explore using a deep convolutional autoencoder for acoustic anomaly detection of wood planers on a new real-life dataset. Specifically, our convolutional autoencoder with skip connections (Skip-CAE) and our Skip-CAE transformer outperform the DCASE autoencoder baseline, one-class SVM, isolation forest and a published convolutional autoencoder architecture, respectively obtaining an area under the ROC curve of 0.846 and 0.875 on a dataset of real-factory planer sounds. Moreover, we show that adding skip connections and attention mechanism under the form of a transformer encoder-decoder helps to further improve the anomaly detection capabilities.

2501.04820

Vasudha Varadarajan

Allison Lahnala and Vasudha Varadarajan and Lucie Flek and H. Andrew Schwartz and Ryan L. Boyd

Unifying the Extremes: Developing a Unified Model for Detecting and Predicting Extremist Traits and Radicalization

17 pages, 7 figures, 4 tables

cs.SI cs.CL cs.CY

http://creativecommons.org/licenses/by/4.0/

The proliferation of ideological movements into extremist factions via social media has become a global concern. While radicalization has been studied extensively within the context of specific ideologies, our ability to accurately characterize extremism in more generalizable terms remains underdeveloped. In this paper, we propose a novel method for extracting and analyzing extremist discourse across a range of online community forums. By focusing on verbal behavioral signatures of extremist traits, we develop a framework for quantifying extremism at both user and community levels. Our research identifies 11 distinct factors, which we term ``The Extremist Eleven,'' as a generalized psychosocial model of extremism. Applying our method to various online communities, we demonstrate an ability to characterize ideologically diverse communities across the 11 extremist traits. We demonstrate the power of this method by analyzing user histories from members of the incel community. We find that our framework accurately predicts which users join the incel community up to 10 months before their actual entry with an AUC of $>0.6$, steadily increasing to AUC ~0.9 three to four months before the event. Further, we find that upon entry into an extremist forum, the users tend to maintain their level of extremism within the community, while still remaining distinguishable from the general online discourse. Our findings contribute to the study of extremism by introducing a more holistic, cross-ideological approach that transcends traditional, trait-specific models.

2501.04821

T. Hakioglu

Susy Exists, Sougata Mardanya, Robert Markiewicz, Tugrul Hakioglu, Jouko Nieminen, Ville J. H\"ark\"onen, Cem Sanga, Arun Bansil, Sugata Chowdhury

Giant Kohn anomaly and chiral phonons in the charge density wave phase of 1H-NbSe$_2$

16 pages, 13 figures

cond-mat.str-el cond-mat.mtrl-sci

http://creativecommons.org/licenses/by-nc-sa/4.0/

Despite extensive investigations, many aspects of charge density waves (CDWs) remain elusive, especially the relative roles of electron-phonon coupling and Fermi surface nesting as the underlying driving mechanisms responsible for the emergence of the CDW vector $\bl Q_{CDW}$. It is puzzling that even though electrons interact strongly with optical phonons in many correlated systems, the actual mode softening is of an acoustic mode. Here we consider monolayer 1H-NbSe$_2$ as an exemplar system, and through an accurate computation of the phonon self-energy, including its off-diagonal components. We provide compelling evidence that the relevant mode is a longitudinal optical phonon that softens by anti-crossing several intervening phonon bands. We also show that $\bl Q_{CDW}$ is fixed by the convolution of the susceptibility and electron-phonon coupling, and that the softened phonons are circularly polarized.

2501.04822

Jeffrey Calhoun

Carina G Biar, Nicholas Bodkin, Gemma L Carvill, Jeffrey D Calhoun

Curated loci prime editing (cliPE) for accessible multiplexed assays of variant effect (MAVEs)

28 pages, 2 figures

q-bio.GN

http://creativecommons.org/licenses/by/4.0/

Multiplexed assays of variant effect (MAVEs) perform simultaneous characterization of many variants. Prime editing has been recently adopted for introducing many variants in their native genomic contexts. However, robust protocols and standards are limited, preventing widespread uptake. Herein, we describe curated loci prime editing (cliPE) which is an accessible, low-cost experimental pipeline to perform MAVEs using prime editing of a target gene, as well as a companion Shiny app (pegRNA Designer) to rapidly and easily design user-specific MAVE libraries.

2501.04823

Aaron Feldman

Aaron O. Feldman, Joseph A. Vincent, Maximilian Adang, Jun En Low, Mac Schwager

Learning Robot Safety from Sparse Human Feedback using Conformal Prediction

cs.RO math.OC stat.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Ensuring robot safety can be challenging; user-defined constraints can miss edge cases, policies can become unsafe even when trained from safe data, and safety can be subjective. Thus, we learn about robot safety by showing policy trajectories to a human who flags unsafe behavior. From this binary feedback, we use the statistical method of conformal prediction to identify a region of states, potentially in learned latent space, guaranteed to contain a user-specified fraction of future policy errors. Our method is sample-efficient, as it builds on nearest neighbor classification and avoids withholding data as is common with conformal prediction. By alerting if the robot reaches the suspected unsafe region, we obtain a warning system that mimics the human's safety preferences with guaranteed miss rate. From video labeling, our system can detect when a quadcopter visuomotor policy will fail to steer through a designated gate. We present an approach for policy improvement by avoiding the suspected unsafe region. With it we improve a model predictive controller's safety, as shown in experimental testing with 30 quadcopter flights across 6 navigation tasks. Code and videos are provided.

2501.04824

Stephane Perrard

Baptiste Auvity, Laurent Duchemin, Antonin Eddi, St\'ephane Perrard

Wave induced fracture of a sea ice analog

physics.flu-dyn

http://creativecommons.org/licenses/by/4.0/

We study at the laboratory scale the rupture of thin floating sheets made of a brittle material under wave induced mechanical forcing. We show that the rupture occurs where the curvature is maximum, and the break up threshold strongly depends on the wave properties. We observe that the corresponding critical stress for fracture depends on the forcing wavelength: our observations are thus incompatible with a critical stress criteria for fracture. Our measurements can rather be rationalized using a energy criteria: a fracture propagates when the material surface energy is lower than the released elastic energy, which depends on the forcing geometry. Our results suggest that current models of sea ice fracture by ocean waves could be refined.

2501.04825

Austin McDannald Ph.D.

Austin McDannald, Daniel W. Siderius, Brian DeCost, Kamal Choudhary, Diana L. Ortiz-Montalvo

Intrinsic Direct Air Capture

cond-mat.mtrl-sci

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We present new metrics to evaluate solid sorbent materials for Direct Air Capture (DAC). These new metrics provide a theoretical upper bound on CO2 captured per energy as well as a theoretical upper limit on the purity of the captured CO2. These new metrics are based entirely on intrinsic material properties and are therefore agnostic to the design of the DAC system. These metrics apply to any adsorption-refresh cycle design. In this work we demonstrate the use of these metrics with the example of temperature-pressure swing refresh cycles. The main requirement for applying these metrics is to describe the equilibrium uptake (along with a few other materials properties) of each species in terms of the thermodynamic variables (e.g. temperature, pressure). We derive these metrics from thermodynamic energy balances. To apply these metrics on a set of examples, we first generated approximations of the necessary materials properties for 11 660 metal-organic framework materials (MOFs). We find that the performance of the sorbents is highly dependent on the path through thermodynamic parameter space. These metrics allow for: 1) finding the optimum materials given a particular refresh cycle, and 2) finding the optimum refresh cycles given a particular sorbent. Applying these metrics to the database of MOFs lead to the following insights: 1) start cold - the equilibrium uptake of CO2 diverges from that of N2 at lower temperatures, and 2) selectivity of CO2 vs other gases at any one point in the cycle does not matter - what matters is the relative change in uptake along the cycle.

2501.04826

Ismail Mustapha Dr

Ismail B. Mustapha, Muyideen Abdulkareem, Shafaatunnur Hasan, Abideen Ganiyu, Hatem Nabus, Jin Chai Lee

Intelligent Gradient Boosting Algorithms for Estimating Strength of Modified Subgrade Soil

17 pages

cs.LG cs.AI cs.CE

http://creativecommons.org/licenses/by/4.0/

The performance of pavement under loading depends on the strength of the subgrade. However, experimental estimation of properties of pavement strengths such as California bearing ratio (CBR), unconfined compressive strength (UCS) and resistance value (R) are often tedious, time-consuming and costly, thereby inspiring a growing interest in machine learning based tools which are simple, cheap and fast alternatives. Thus, the potential application of two boosting techniques; categorical boosting (CatBoost) and extreme gradient boosting (XGBoost) and support vector regression (SVR), is similarly explored in this study for estimation of properties of subgrade soil modified with hydrated lime activated rice husk ash (HARSH). Using 121 experimental data samples of varying proportions of HARSH, plastic limit, liquid limit, plasticity index, clay activity, optimum moisture content, and maximum dry density as input for CBR, UCS and R estimation, four evaluation metrics namely coefficient of determination (R2), root mean squared error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE) are used to evaluate the models' performance. The results indicate that XGBoost outperformed CatBoost and SVR in estimating these properties, yielding R2 of 0.9994, 0.9995 and 0.9999 in estimating the CBR, UCS and R respectively. Also, SVR outperformed CatBoost in estimating the CBR and R with R2 of 0.9997 respectively. On the other hand, CatBoost outperformed SVR in estimating the UCS with R2 of 0.9994. Feature sensitivity analysis shows that the three machine learning techniques are unanimous that increasing HARSH proportion lead to values of the estimated properties respectively. A comparison with previous results also shows superiority of XGBoost in estimating subgrade properties.

2501.04827

Jacob Notbohm

Pradip K. Bera, Molly McCord, Jun Zhang, Jacob Notbohm

Energy Dynamics Powered by Traction and Stress Control Formation and Motion of +1/2 Topological Defects in Epithelial Cell Monolayers

physics.bio-ph cond-mat.soft

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In confluent cell monolayers, patterns of cell forces and motion are systematically altered near topological defects in cell shape. In turn, defects have been proposed to alter cell density, extrusion, and invasion, but it remains unclear how the defects form and how they affect cell forces and motion. Here, we studied +1/2 defects, and, in contrast to prior studies, we observed both tail-to-head and head-to-tail defect motion occurring at the same time in the same cell monolayer. We quantified the cell velocities, the tractions at the cell-substrate interface, and stresses within the cell layer near +1/2 defects. Results revealed that both traction and stress are sources of activity within the epithelial cell monolayer, with their competition defining whether the cells inject or dissipate energy and determining the direction of motion of +1/2 defects. Interestingly, patterns of motion, traction, stress, and energy injection near +1/2 defects existed before defect formation, suggesting that defects form as a result of spatially coordinated patterns in cell forces and motion. These findings reverse the current picture, from one in which defects define the cell forces and motion to one in which coordinated patterns of cell forces and motion cause defects to form and move.

2501.04828

\c{S}aziye Bet\"ul \"Ozate\c{s}

\c{S}aziye Bet\"ul \"Ozate\c{s}, Tar{\i}k Emre T{\i}ra\c{s}, Ece Elif Adak, Berat Do\u{g}an, Fatih Burak Karag\"oz, Efe Eren Gen\c{c}, Esma F. Bilgin Ta\c{s}demir

Building Foundations for Natural Language Processing of Historical Turkish: Resources and Models

cs.CL

http://creativecommons.org/licenses/by-nc-nd/4.0/

This paper introduces foundational resources and models for natural language processing (NLP) of historical Turkish, a domain that has remained underexplored in computational linguistics. We present the first named entity recognition (NER) dataset, HisTR and the first Universal Dependencies treebank, OTA-BOUN for a historical form of the Turkish language along with transformer-based models trained using these datasets for named entity recognition, dependency parsing, and part-of-speech tagging tasks. Additionally, we introduce Ottoman Text Corpus (OTC), a clean corpus of transliterated historical Turkish texts that spans a wide range of historical periods. Our experimental results show significant improvements in the computational analysis of historical Turkish, achieving promising results in tasks that require understanding of historical linguistic structures. They also highlight existing challenges, such as domain adaptation and language variations across time periods. All of the presented resources and models are made available at https://huggingface.co/bucolin to serve as a benchmark for future progress in historical Turkish NLP.

2501.04829

Antonina Shibalova

V.N. Obridko, A.S. Shibalova, D.D. Sokoloff

Structures of various types of symmetry in the solar activity cycle

18 pages, 8 figures

astro-ph.SR

http://creativecommons.org/licenses/by/4.0/

The solar cycle is a complex phenomenon. To comprehensively understand it, we have to study various tracers. The most important component of this complex is the solar dynamo, which is understood as self-excitation of the solar magnetic field in the form of traveling waves somewhere in the convection zone. Along with the solar dynamo, the formation of the solar cycle involves other processes that are associated with the dynamo but are not its necessary part. We give a review of such phenomena that have not yet been explained in terms of dynamo theory. We consider the manifestations of the solar cycle in harmonics of the solar large-scale surface magnetic field, including zonal, sectorial, and tesseral harmonics; analyze their contribution to magnetic energy; and identify phases of the activity cycle using harmonics of different types of symmetry. The universal magnetic scenario of a solar activity cycle does not depend on its number and height. At the beginning of the cycle on the photosphere, the zonal harmonics account for 37-42% of the total energy (not 100%, as assumed in simplified descriptions). Sectorial harmonics do not disappear at all but account for 5-10% of the total energy. At this stage, the greatest energy (about 40%) is contained in the tesseral harmonics. As the cycle develops, the relative energy of zonal harmonics gradually decreases, reaching a minimum of 15-18% immediately before the onset of the sunspot maximum. The relative energy of sectorial harmonics increases and reaches a maximum (60-65%) somewhat later than the calendar date of the sunspot maximum. A particular feature of the tesseral harmonics is that their relative energy index changes in a much narrower range and never falls below 40% even at the cycle minimum. This is due to active regions and nonglobal magnetic fields. It is possible that tesseral harmonics are formed in shallow subphotospheric layers.

2501.04830

Xuesong Wang

Xuesong Wang and Caisheng Wang

A Deep Learning-Based Method for Power System Resilience Evaluation

Submitted to IEEE Transactions on Power Systems

eess.SY cs.SY

http://creativecommons.org/licenses/by/4.0/

Power systems are critical infrastructure in modern society, and power outages can cause significant disruptions to communities and individuals' daily lives. The resilience of a power system measures its ability to maintain power supply during highly disruptive events such as hurricanes, earthquakes, and thunderstorms. Traditional methods for quantifying power system resilience include statistics-based and simulation-based approaches. Statistics-based methods offer a retrospective analysis of system performance without requiring a physical model, while simulation-based methods necessitate detailed physical system information and often simplify real-world scenarios. This paper introduces a deep learning-based method for evaluating power system resilience using historical power outage data. The method leverages the generalization capabilities of deep learning models and incorporates socio-economic and demographic factors as weighting terms to highlight the impacts on vulnerable demographic groups. The effectiveness of the proposed method is demonstrated through two case studies: one with real historical outage data and the other with simulated outage records. This approach provides valuable insights into measuring power system resilience against hazardous weather events without requiring a physical model of the target systems. The evaluation results can further guide the planning of distributed energy resources for resilience enhancement.

2501.04831

Himanshu Thapliyal

Md Saif Hassan Onim, Travis S. Humble and Himanshu Thapliyal

Quantum Hybrid Support Vector Machines for Stress Detection in Older Adults

quant-ph cs.LG

http://creativecommons.org/licenses/by-nc-sa/4.0/

Stress can increase the possibility of cognitive impairment and decrease the quality of life in older adults. Smart healthcare can deploy quantum machine learning to enable preventive and diagnostic support. This work introduces a unique technique to address stress detection as an anomaly detection problem that uses quantum hybrid support vector machines. With the help of a wearable smartwatch, we mapped baseline sensor reading as normal data and stressed sensor reading as anomaly data using cortisol concentration as the ground truth. We have used quantum computing techniques to explore the complex feature spaces with kernel-based preprocessing. We illustrate the usefulness of our method by doing experimental validation on 40 older adults with the help of the TSST protocol. Our findings highlight that using a limited number of features, quantum machine learning provides improved accuracy compared to classical methods. We also observed that the recall value using quantum machine learning is higher compared to the classical method. The higher recall value illustrates the potential of quantum machine learning in healthcare, as missing anomalies could result in delayed diagnostics or treatment.

2501.04832

Benjamin Goertzel

Ben Goertzel

ActPC-Geom: Towards Scalable Online Neural-Symbolic Learning via Accelerating Active Predictive Coding with Information Geometry & Diverse Cognitive Mechanisms

cs.AI cs.LG cs.NE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper introduces ActPC-Geom, an approach to accelerate Active Predictive Coding (ActPC) in neural networks by integrating information geometry, specifically using Wasserstein-metric-based methods for measure-dependent gradient flows. We propose replacing KL-divergence in ActPC's predictive error assessment with the Wasserstein metric, suggesting this may enhance network robustness. To make this computationally feasible, we present strategies including: (1) neural approximators for inverse measure-dependent Laplacians, (2) approximate kernel PCA embeddings for low-rank approximations feeding into these approximators, and (3) compositional hypervector embeddings derived from kPCA outputs, with algebra optimized for fuzzy FCA lattices learned through neural architectures analyzing network states. This results in an ActPC architecture capable of real-time online learning and integrating continuous (e.g., transformer-like or Hopfield-net-like) and discrete symbolic ActPC networks, including frameworks like OpenCog Hyperon or ActPC-Chem for algorithmic chemistry evolution. Shared probabilistic, concept-lattice, and hypervector models enable symbolic-subsymbolic integration. Key features include (1) compositional reasoning via hypervector embeddings in transformer-like architectures for tasks like commonsense reasoning, and (2) Hopfield-net dynamics enabling associative long-term memory and attractor-driven cognitive features. We outline how ActPC-Geom combines few-shot learning with online weight updates, enabling deliberative thinking and seamless symbolic-subsymbolic reasoning. Ideas from Galois connections are explored for efficient hybrid ActPC/ActPC-Chem processing. Finally, we propose a specialized HPC design optimized for real-time focused attention and deliberative reasoning tailored to ActPC-Geom's demands.

2501.04833

Zehui Liu

Zehui Liu, Qingsong Wang, Chunfeng Cui

Multi-step Inertial Accelerated Doubly Stochastic Gradient Methods for Block Term Tensor Decomposition

math.OC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we explore a specific optimization problem that combines a differentiable nonconvex function with a nondifferentiable function for multi-block variables, which is particularly relevant to tackle the multilinear rank-($L_r$,$L_r$,1) block-term tensor decomposition model with a regularization term. While existing algorithms often suffer from high per-iteration complexity and slow convergence, this paper employs a unified multi-step inertial accelerated doubly stochastic gradient descent method tailored for structured rank-$\left(L_r, L_r, 1\right)$ tensor decomposition, referred to as Midas-LL1. We also introduce an extended multi-step variance-reduced stochastic estimator framework. Our analysis under this new framework demonstrates the subsequential and sequential convergence of the proposed algorithm under certain conditions and illustrates the sublinear convergence rate of the subsequence, showing that the Midas-LL1 algorithm requires at most $\mathcal{O}(\varepsilon^{-2})$ iterations in expectation to reach an $\varepsilon$-stationary point. The proposed algorithm is evaluated on several datasets, and the results indicate that Midas-LL1 outperforms existing state-of-the-art algorithms in terms of both computational speed and solution quality.

2501.04834

Riley Rosener

Riley Rosener, Michael Zhang, and Jacob L. Bean

Detectability of Emission from Exoplanet Outflows Calculated by pyTPCI, a New 1D Radiation-Hydrodynamic Code

Accepted for publication in The Astrophysical Journal. 17 pages, 8 figures, 4 tables

10.3847/1538-4357/ada6ab

astro-ph.EP astro-ph.IM

http://creativecommons.org/licenses/by-nc-sa/4.0/

Photoevaporation in exoplanet atmospheres is thought to contribute to the shaping of the small planet radius valley. Escaping atmospheres have been detected in transmission across a variety of exoplanet types, from hot Jupiters to mini-Neptunes. However, no work has yet considered whether outflows might also be detectable in emission. We introduce pyTPCI, a new, open-source self-consistent 1D radiative-hydrodynamics code that is an improved version of The PLUTO-CLOUDY Interface. We use pyTPCI to model seven exoplanets (HD 189733b, HD 209458b, WASP-69b, WASP-107b, TOI-1430b, TOI-560b, and HAT-P-32b) at varying metallicities and compute their emission spectra to investigate their detectability across a variety of spectral lines. We calculate the eclipse depths and signal-to-noise ratios (SNR) of these lines for a 10m class telescope with a high-resolution spectrograph, taking into account appropriate line broadening mechanisms. We show that the most detectable spectral lines tend to be the 589 nm Na I doublet and the 1083 nm metastable helium triplet. Halpha and Mg I 457 nm are moderately strong for some planets at some metallicities, but they are almost always optically thin, so some of their emission may not be from the outflow. The planet with the highest-flux, highest-eclipse-depth, and highest-SNR lines is HD 189733b, with a Na I eclipse depth of 410 ppm and SNR of 2.4 per eclipse, and a He* eclipse depth of 170 ppm and SNR of 1.3. These signals would be marginally detectable with Keck if 3-10 eclipses were observed, assuming (over-optimistically) photon limited observations.

2501.04835

Zhenyu Pan

Zhenyu Pan, Xuefeng Song, Yunkun Wang, Rongyu Cao, Binhua Li, Yongbin Li, Han Liu

Do Code LLMs Understand Design Patterns?

accpeted by llm4code workshop in ICSE 2025

cs.SE cs.AI

http://creativecommons.org/licenses/by-nc-sa/4.0/

Code Large Language Models (LLMs) demonstrate great versatility in adapting to various downstream tasks, including code generation and completion, as well as bug detection and fixing. However, Code LLMs often fail to capture existing coding standards, leading to the generation of code that conflicts with the required design patterns for a given project. As a result, developers must post-process to adapt the generated code to the project's design norms. In this work, we empirically investigate the biases of Code LLMs in software development. Through carefully designed experiments, we assess the models' understanding of design patterns across recognition, comprehension, and generation. Our findings reveal that biases in Code LLMs significantly affect the reliability of downstream tasks.

2501.04836

Artuur Stevens

Artuur Stevens and Christophe Caloz

Photon State Evolution in Arbitrary Time-Varying Media

11 pages, 2 figures

physics.optics cond-mat.mes-hall cond-mat.mtrl-sci quant-ph

http://creativecommons.org/licenses/by/4.0/

We introduce the instantaneous eigenstate method to study the evolution of quantum states in media with arbitrary time-varying permittivity and permeability. This method leverages the Heisenberg equation to bypass the Schr\"odinger equation, which leads to a complicated infinite set of coupled differential equations. Instead, the method allows the computation of the state evolution by solving only two coupled differential equations. Using this approach, we draw general conclusions about photon statistics in time-varying media. Our findings reveal that the maximum probability of generating a single photon pair from vacuum in such media is $25\%$, while Bell states can be created with a maximum probability of $84\%$. Additionally, we demonstrate that the spectral profile of emitted photons can be precisely controlled through the temporal profiles of permittivity and permeability. These results provide deeper insights into photon state manipulation in time-varying media. Furthermore, the instantaneous eigenstate method opens new opportunities to study state evolution in other systems where the Heisenberg equation offers a more tractable solution than the Schr\"odinger equation.

2501.04837

Alberto Carnelli

Carnelli Alberto

Proposal for simplified template cross-sections extension using $\cal{CP}$ observables in $t\overline{t}H$

Poster at the 17th International Workshop on Top Quark Physics (Top2024), 22-27 September 2024

hep-ph

http://creativecommons.org/licenses/by/4.0/

The Large Hadron Collider (LHC) offers a unique opportunity to investigate $\cal{CP}$ violation in the Yukawa coupling between the Higgs boson and the top quark by studying Higgs production in association with top quarks; this is of fundamental importance, seeing that the $\cal{CP}$ properties of the Higgs boson are yet to measure with high precision. To address this, the focus of this work has been an extension of the simplified template cross-section (STXS) framework, devised to be sensitive to $\cal{CP}$ effects. Our study focused on $\cal{CP}$-sensitive observables across multiple Higgs decay channels, comparing their performances. The result indicates that the most efficient extension of the current binning used in the STXS framework, which currently uses the Higgs boson's transverse momentum $p_{T,H}$, requires adding one further split using $\cal{CP}$-sensitive observables. Between these observables, one of the best is the Collins-Soper angle $|\cos\theta^*|$, a variable derived from momenta information of the top quarks. We have investigated the improvement brought by our two-dimensional STXS setup and compared it to the currently employed methodologies, finding an increase in performances at an integrated luminosity of $300$ fb$^{-1}$. Moreover, our results highlight that this advantage seems to be present also at $3000$ fb$^{-1}$.

2501.04838

Savannah Gowen

Savannah D. Gowen

Training Allostery-Inspired Mechanical Response in Disordered Elastic Networks

cond-mat.soft

http://creativecommons.org/licenses/by/4.0/

Disordered elastic networks are a model material system in which it is possible to achieve tunable and trainable functions. This work investigates the modification of local mechanical properties in disordered networks inspired by allosteric interactions in proteins: applying strain locally to a set of source nodes triggers a strain response at a distant set of target nodes. This is demonstrated first by using directed aging to modify the existing mechanical coupling between pairs of distant source and target nodes, and later as a means for inducing coupling between formerly isolated source-target pairs. The experimental results are compared with those predicted by simulations.

2501.04839

Xianbang Chen Mr.

Xianbang Chen, Yikui Liu, Neng Fan, Lei Wu

DRL-Based Medium-Term Planning of Renewable-Integrated Self-Scheduling Cascaded Hydropower to Guide Wholesale Market Participation

eess.SY cs.SY

http://creativecommons.org/licenses/by/4.0/

For self-scheduling cascaded hydropower (S-CHP) facilities, medium-term planning is a critical step that coordinates water availability over the medium-term horizon, providing water usage guidance for their short-term operations in wholesale market participation. Typically, medium-term planning strategies (e.g., reservoir storage targets at the end of each short-term period) are determined by either optimization methods or rules of thumb. However, with the integration of variable renewable energy sources (VRESs), optimization-based methods suffer from deviations between the anticipated and actual reservoir storage, while rules of thumb could be financially conservative, thereby compromising short-term operating profitability in wholesale market participation. This paper presents a deep reinforcement learning (DRL)-based framework to derive medium-term planning policies for VRES-integrated S-CHPs (VS-CHPs), which can leverage contextual information underneath individual short-term periods and train planning policies by their induced short-term operating profits in wholesale market participation. The proposed DRL-based framework offers two practical merits. First, its planning strategies consider both seasonal requirements of reservoir storage and needs for short-term operating profits. Second, it adopts a multi-parametric programming-based strategy to accelerate the expensive training process associated with multi-step short-term operations. Finally, the DRL-based framework is evaluated on a real-world VS-CHP, demonstrating its advantages over current practice.

2501.04840

Alessandro Nicolaou AN

Alessandro Nicolaou, Kari Ruotsalainen, Laura Susana, Victor Por\'ee, Luiz Galvao Tizei, Jaakko Koskelo, Takashi Taniguchi, Kenji Watanabe, Alberto Zobelli and Matteo Gatti

Direct measurement of the longitudinal exciton dispersion in hBN by resonant inelastic x-ray scattering

cond-mat.mtrl-sci cond-mat.str-el

http://creativecommons.org/licenses/by/4.0/

We report resonant inelastic X-ray scattering (RIXS) measurements on the prototypical hexagonal boron nitride hBN layered compound. The RIXS results at the B and N K edges have been combined with electron energy loss spectroscopy (EELS) experiments and ab initio calculations within the framework of the Bethe-Salpeter equation of many-body perturbation theory. By means of this tight interplay of different spectroscopies, the lowest longitudinal exciton of hBN has been identified. Moreover, its qualitatively different dispersions along the $\Gamma$K and the $\Gamma$M directions of the Brillouin zone have been determined. Our study advocates soft X-ray RIXS and EELS to be a promising combination to investigate electronic excitations in materials.

2501.04841

Ka Wai Wu

Ka Wai Wu

Blockchain-Based Secure Vehicle Auction System with Smart Contracts

cs.CR

http://creativecommons.org/licenses/by/4.0/

The problem of a single point of failure in centralized systems poses a great challenge to the stability of such systems. Meanwhile, the tamperability of data within centralized systems makes users reluctant to trust and use centralized applications in many scenarios, including the financial and business sectors. Blockchain, as a new decentralized technology, addresses these issues effectively. As a typical decentralized system, blockchain can be utilized to build a data-sharing model. Users in a blockchain do not need to trust other users; instead, they trust that the majority of miner nodes are honest. Smart contracts enable developers to write distributed programs based on blockchain systems, ensuring that all code is immutable and secure. In this paper, we analyze the security of blockchain technology to illustrate its advantages and justify its use. Furthermore, we design a new system for storing and trading vehicle information based on the Ethereum blockchain and smart contract technology. Specifically, our system allows users to upload vehicle information and auction vehicles to transfer ownership. Our application provides great convenience to buyers and owners, while the use of smart contracts enhances the security and privacy of the system.

2501.04842

Nicolas Chopin

Nicolas Chopin, Hejin Wang, Mathieu Gerber

Adaptive stratified Monte Carlo using decision trees

20 pages, 6 figures

stat.CO

http://creativecommons.org/licenses/by/4.0/

It has been known for a long time that stratification is one possible strategy to obtain higher convergence rates for the Monte Carlo estimation of integrals over the hyper-cube $[0, 1]^s$ of dimension $s$. However, stratified estimators such as Haber's are not practical as $s$ grows, as they require $\mathcal{O}(k^s)$ evaluations for some $k\geq 2$. We propose an adaptive stratification strategy, where the strata are derived from a a decision tree applied to a preliminary sample. We show that this strategy leads to higher convergence rates, that is, the corresponding estimators converge at rate $\mathcal{O}(N^{-1/2-r})$ for some $r>0$ for certain classes of functions. Empirically, we show through numerical experiments that the method may improve on standard Monte Carlo even when $s$ is large.

2501.04843

Henrik Christiansen

Henrik Christiansen, Suman Majumder, Wolfhard Janke, Malte Henkel

Finite-Size Effects in Aging can be Interpreted as Sub-Aging

cond-mat.stat-mech physics.comp-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Systems brought out of equilibrium through a rapid quench from a disordered initial state into an ordered phase undergo physical aging in the form of phase-ordering kinetics, with characteristic dynamical scaling. In many systems, notably glasses, dynamical scaling is often described through sub-aging, where a phenomenological sub-aging exponent $0<\mu< 1$ is empirically chosen to achieve the best possible data collapse. Here it is shown that finite-size effects modify the dynamical scaling behavior, away from simple aging with $\mu=1$ towards $\mu<1$, such that phenomenologically it would appear as sub-aging. This is exemplified for the exactly solved dynamical spherical model in dimensions $2<d<4$ and numerical simulations of the two-dimensional Ising model, with short-ranged and long-ranged interactions.

2501.04844

Jihwan Lee

Jihwan Lee, Tiantian Feng, Aditya Kommineni, Sudarsana Reddy Kadiri, Shrikanth Narayanan

Enhancing Listened Speech Decoding from EEG via Parallel Phoneme Sequence Prediction

ICASSP 2025

eess.AS cs.AI cs.CL eess.SP

http://creativecommons.org/licenses/by/4.0/

Brain-computer interfaces (BCI) offer numerous human-centered application possibilities, particularly affecting people with neurological disorders. Text or speech decoding from brain activities is a relevant domain that could augment the quality of life for people with impaired speech perception. We propose a novel approach to enhance listened speech decoding from electroencephalography (EEG) signals by utilizing an auxiliary phoneme predictor that simultaneously decodes textual phoneme sequences. The proposed model architecture consists of three main parts: EEG module, speech module, and phoneme predictor. The EEG module learns to properly represent EEG signals into EEG embeddings. The speech module generates speech waveforms from the EEG embeddings. The phoneme predictor outputs the decoded phoneme sequences in text modality. Our proposed approach allows users to obtain decoded listened speech from EEG signals in both modalities (speech waveforms and textual phoneme sequences) simultaneously, eliminating the need for a concatenated sequential pipeline for each modality. The proposed approach also outperforms previous methods in both modalities. The source code and speech samples are publicly available.

2501.04845

Jakub Kvapil

J. Kvapil (1), G. Borca-Tasciuc (2), H. Bossi (3), K. Chen (4), Y. Chen (4), Y. Corrales Morales (3), H. Da Costa (1), C. Da Silva (1), C. Dean (3), J. Durham (1), S. Fu (5), C. Hao (6), P. Harris (3), O. Hen (3), H. Jheng (3), Y. Lee (3), P. Li (6), X. Li (1), Y. Lin (1), M. X. Liu (1), V. Loncar (3), J. P. Mitrevski (8), A. Olvera (5), M. L. Purschke (7), J. S. Renck (1), G. Roland (3), J. Schambach (9), Z. Shi (1), N. Tran (8), N. Wuerfel (10), B. Xu (6), D. Yu (11), H. Zhang (6) ((1) Los Alamos National Laboratory, (2) Rensselaer Polytechnic Institute, (3) Massachusetts Institute of Technology, (4) Central China Normal University, (5) University of North Texas, (6) Georgia Institute of Technology, (7) Brookhaven National Laboratory, (8) Fermilab, (9) Oak Ridge National Laboratory, (10) University of Michigan, (11) New Jersey Institute of Technology)

Intelligent experiments through real-time AI: Fast Data Processing and Autonomous Detector Control for sPHENIX and future EIC detectors

proceedings for 42nd International Conference on High Energy Physics (ICHEP2024), 18-24 July 2024, Prague, Czech Republic

LA-UR-24-30394

physics.ins-det cs.LG hep-ex nucl-ex

http://creativecommons.org/licenses/by/4.0/

This R\&D project, initiated by the DOE Nuclear Physics AI-Machine Learning initiative in 2022, leverages AI to address data processing challenges in high-energy nuclear experiments (RHIC, LHC, and future EIC). Our focus is on developing a demonstrator for real-time processing of high-rate data streams from sPHENIX experiment tracking detectors. The limitations of a 15 kHz maximum trigger rate imposed by the calorimeters can be negated by intelligent use of streaming technology in the tracking system. The approach efficiently identifies low momentum rare heavy flavor events in high-rate p+p collisions (3MHz), using Graph Neural Network (GNN) and High Level Synthesis for Machine Learning (hls4ml). Success at sPHENIX promises immediate benefits, minimizing resources and accelerating the heavy-flavor measurements. The approach is transferable to other fields. For the EIC, we develop a DIS-electron tagger using Artificial Intelligence - Machine Learning (AI-ML) algorithms for real-time identification, showcasing the transformative potential of AI and FPGA technologies in high-energy nuclear and particle experiments real-time data processing pipelines.

2501.04846

Yachuan Li

Yachuan Li, Xavier Soria Poma, Yun Bai, Qian Xiao, Chaozhi Yang, Guanlin Li, Zongmin Li

EDMB: Edge Detector with Mamba

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Transformer-based models have made significant progress in edge detection, but their high computational cost is prohibitive. Recently, vision Mamba have shown excellent ability in efficiently capturing long-range dependencies. Drawing inspiration from this, we propose a novel edge detector with Mamba, termed EDMB, to efficiently generate high-quality multi-granularity edges. In EDMB, Mamba is combined with a global-local architecture, therefore it can focus on both global information and fine-grained cues. The fine-grained cues play a crucial role in edge detection, but are usually ignored by ordinary Mamba. We design a novel decoder to construct learnable Gaussian distributions by fusing global features and fine-grained features. And the multi-grained edges are generated by sampling from the distributions. In order to make multi-granularity edges applicable to single-label data, we introduce Evidence Lower Bound loss to supervise the learning of the distributions. On the multi-label dataset BSDS500, our proposed EDMB achieves competitive single-granularity ODS 0.837 and multi-granularity ODS 0.851 without multi-scale test or extra PASCAL-VOC data. Remarkably, EDMB can be extended to single-label datasets such as NYUDv2 and BIPED. The source code is available at https://github.com/Li-yachuan/EDMB.

2501.04847

Sepehrdad Tahmasebi

Sepehrdad Tahmasebi, Geng Tian, Shaoxiang Qin, Ahmed Marey, Liangzhu Leon Wang, Saeed Rayegan

Using Diffusion Models for Reducing Spatiotemporal Errors of Deep Learning Based Urban Microclimate Predictions at Post-Processing Stage

physics.flu-dyn

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Computational fluid dynamics (CFD) is a powerful tool for modeling turbulent flow and is commonly used for urban microclimate simulations. However, traditional CFD methods are computationally intensive, requiring substantial hardware resources for high-fidelity simulations. Deep learning (DL) models are becoming popular as efficient alternatives as they require less computational resources to model complex non-linear interactions in fluid flow simulations. A major drawback of DL models is that they are prone to error accumulation in long-term temporal predictions, often compromising their accuracy and reliability. To address this shortcoming, this study investigates the use of a denoising diffusion probabilistic model (DDPM) as a novel post-processing technique to mitigate error propagation in DL models' sequential predictions. To address this, we employ convolutional autoencoder (CAE) and U-Net architectures to predict airflow dynamics around a cubic structure. The DDPM is then applied to the models' predictions, refining the reconstructed flow fields to better align with high-fidelity statistical results obtained from large-eddy simulations. Results demonstrate that, although deep learning models provide significant computational advantages over traditional numerical solvers, they are susceptible to error accumulation in sequential predictions; however, utilizing DDPM as a post-processing step enhances the accuracy of DL models by up to 65% while maintaining a 3 times speedup compared to traditional numerical solvers. These findings highlight the potential of integrating denoising diffusion probabilistic models as a transformative approach to improving the reliability and accuracy of deep learning-based urban microclimate simulations, paving the way for more efficient and scalable fluid dynamics modeling.

2501.04848

Aisha Ali-Gombe

Brandon J Walton, Mst Eshita Khatun, James M Ghawaly, Aisha Ali-Gombe

Exploring Large Language Models for Semantic Analysis and Categorization of Android Malware

Workshop on AI for Cyber Threat Intelligence (WAITI) 2024

cs.CR cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Malware analysis is a complex process of examining and evaluating malicious software's functionality, origin, and potential impact. This arduous process typically involves dissecting the software to understand its components, infection vector, propagation mechanism, and payload. Over the years, deep reverse engineering of malware has become increasingly tedious, mainly due to modern malicious codebases' fast evolution and sophistication. Essentially, analysts are tasked with identifying the elusive needle in the haystack within the complexities of zero-day malware, all while under tight time constraints. Thus, in this paper, we explore leveraging Large Language Models (LLMs) for semantic malware analysis to expedite the analysis of known and novel samples. Built on GPT-4o-mini model, \msp is designed to augment malware analysis for Android through a hierarchical-tiered summarization chain and strategic prompt engineering. Additionally, \msp performs malware categorization, distinguishing potential malware from benign applications, thereby saving time during the malware reverse engineering process. Despite not being fine-tuned for Android malware analysis, we demonstrate that through optimized and advanced prompt engineering \msp can achieve up to 77% classification accuracy while providing highly robust summaries at functional, class, and package levels. In addition, leveraging the backward tracing of the summaries from package to function levels allowed us to pinpoint the precise code snippets responsible for malicious behavior.

2501.04849

Paul M. Alsing

Paul M. Alsing and Richard J. Birrittella

An examination of the extended Hong-Ou-Mandel effect and considerations for experimental detection

14 pages, 5 figures. arXiv admin note: substantial text overlap with arXiv:2410.11800

quant-ph

http://creativecommons.org/licenses/by-nc-sa/4.0/

In recent works we have explored a multi-photon extension of the celebrated two-photon Hong-Ou-Mandel (HOM) effect in which the quantum amplitudes for a two-photon input to a lossless, balanced 50:50 beamsplitter (BS) undergoes complete destructive interference. In the extended Hong-Ou-Mandel (eHOM) effect the multi-photon scattering of photons from the two input ports to the two output ports of the BS for Fock number basis input states (FS) $|n,m\rangle_{12}$ exhibit complete destructive interference pairwise within the quantum amplitudes containing many scattering components, generalizing the two-photon HOM effect. This has profound implications for arbitrary bipartite photonic input states constructed from such basis states: if the input state to one input port of the BS is of odd parity, i.e. constructed from only of odd numbers of photons, then regardless of the input state to the second 50:50 BS port, there will be a central nodal line (CNL) of zeros in the joint output probability distribution along the main diagonal for coincidence detection. The first goal of this present work is to show diagrammatically how the extended HOM effect can be seen as a succession of multi-photon HOM effects when the latter is viewed as a pairwise cancellation of mirror image scattering amplitudes. The second goal of this work is to explore considerations for the experimental realization of the extended Hong-Ou-Mandel effect. We examine the case of a single photon interfering with a coherent state (an idealized laser) on a balanced 50:50 beamsplitter and consider prospects for experimental detection of the output destructive interference by including additional effects such as imperfect detection efficiency, spatio-temporal mode functions, and time delay between the detected output photons.

2501.04850

Zhixiang Wu

Matthias Strauch, Zhixiang Wu

Bernstein-Zelevinsky duality for locally analytic principal series

46 pages, comments welcome!

math.RT math.NT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We consider certain dual of the Kohlhaase-Schraen resolutions for locally analytic principal series representations of $p$-adic Lie groups in the case of integral weights. The dual complexes calculate the expected Bernstein-Zelevinsky dual of the locally analytic representations and lead to the Grothendieck-Serre duality of coherent sheaves on patched eigenvarieties.

2501.04851

Dan Ismailescu

Dan Ismailescu and Yunkyu James Lee

Polynomially growing integer sequences all whose terms are composite

12 pages, 1 table

math.NT

http://creativecommons.org/licenses/by/4.0/

We identify pairs of positive integers $(t, d)$ with the property that the integer sequence with general term $\lfloor{n^t/d\rfloor}$ contains at most finitely many primes.

2501.04852

Youssef Ahendouz

Youssef Ahendouz and Ismail Akharraz

Classification of Self-Dual Constacyclic Codes of Prime Power Length $p^s$ Over $\frac{\mathbb{F}_{p^m}[u]}{\left\langle u^3\right\rangle} $

cs.IT math.IT math.RA

http://creativecommons.org/licenses/by-sa/4.0/

Let $\mathbb{F}_{p^m}$ be a finite field of cardinality $p^m$, where $p$ is a prime number and $m$ is a positive integer. Self-dual constacyclic codes of length \( p^s \) over \( \frac{\mathbb{F}_{p^m}[u]}{\langle u^3 \rangle} \) exist only when \( p = 2 \). In this work, we classify and enumerate all self-dual cyclic codes of length \( 2^s \) over \( \frac{\mathbb{F}_{2^m}[u]}{\langle u^3 \rangle} \), thereby completing the classification and enumeration of self-dual constacyclic codes of length \( p^s \) over \( \frac{\mathbb{F}_{p^m}[u]}{\langle u^3 \rangle} \). Additionally, we correct and improve results from B. Kim and Y. Lee (2020) in \cite{kim2020classification}.

2501.04853

Akanksha Negi

Akanksha Negi, Didier Nibbering

Identification of dynamic treatment effects when treatment histories are partially observed

econ.EM

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper proposes a class of methods for identifying and estimating dynamic treatment effects when outcomes depend on the entire treatment path and treatment histories are only partially observed. We advocate for the approach which we refer to as `robust' that identifies path-dependent treatment effects for different mover subpopulations under misspecification of any one of three models involved (outcome, propensity score, or missing data models). Our approach can handle fixed, absorbing, sequential, or simultaneous treatment regimes where missing treatment histories may obfuscate identification of causal effects. Numerical experiments demonstrate how the proposed estimator compares to traditional complete-case methods. We find that the missingness-adjusted estimates have negligible bias compared to their complete-case counterparts. As an illustration, we apply the proposed class of adjustment methods to estimate dynamic effects of COVID-19 on voter turnout in the 2022 U.S. general elections. We find that counties that experienced above-average number of cases in 2020 and 2021 had a statistically significant reduction in voter turnout compared to those that did not.

2501.04854

Fernando Granha Jeronimo

Leonardo Nagami Coregliano, Fernando Granha Jeronimo, Chris Jones, Nati Linial and Elyassaf Loyfer

Higher-order Delsarte Dual LPs: Lifting, Constructions and Completeness

50 pages

cs.IT cs.DM math.CO math.IT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

A central and longstanding open problem in coding theory is the rate-versus-distance trade-off for binary error-correcting codes. In a seminal work, Delsarte introduced a family of linear programs establishing relaxations on the size of optimum codes. To date, the state-of-the-art upper bounds for binary codes come from dual feasible solutions to these LPs. Still, these bounds are exponentially far from the best-known existential constructions. Recently, hierarchies of linear programs extending and strengthening Delsarte's original LPs were introduced for linear codes, which we refer to as higher-order Delsarte LPs. These new hierarchies were shown to provably converge to the actual value of optimum codes, namely, they are complete hierarchies. Therefore, understanding them and their dual formulations becomes a valuable line of investigation. Nonetheless, their higher-order structure poses challenges. In fact, analysis of all known convex programming hierarchies strengthening Delsarte's original LPs has turned out to be exceedingly difficult and essentially nothing is known, stalling progress in the area since the 1970s. Our main result is an analysis of the higher-order Delsarte LPs via their dual formulation. Although quantitatively, our current analysis only matches the best-known upper bounds, it shows, for the first time, how to tame the complexity of analyzing a hierarchy strengthening Delsarte's original LPs. In doing so, we reach a better understanding of the structure of the hierarchy, which may serve as the foundation for further quantitative improvements. We provide two additional structural results for this hierarchy. First, we show how to \emph{explicitly} lift any feasible dual solution from level $k$ to a (suitable) larger level $\ell$ while retaining the objective value. Second, we give a novel proof of completeness using the dual formulation.

2501.04855

Xuan Thang Duong

Thang Xuan Duong, Farshad Roohbakhshan and Roger Andrew Sauer

A new rotation-free isogeometric thin shell formulation and a corresponding continuity constraint for patch boundaries

This is a corrected version of the published journal article. All corrections are marked in green. They were typographical errors that did not affect simulation results

Comput. Methods Appl. Mech. Engrg., Vol 316, P43-83 (2017)

10.1016/j.cma.2016.04.008

cs.CE

http://creativecommons.org/licenses/by-nc-sa/4.0/

This paper presents a general non-linear computational formulation for rotation-free thin shells based on isogeometric finite elements. It is a displacement-based formulation that admits general material models. The formulation allows for a wide range of constitutive laws, including both shell models that are extracted from existing 3D continua using numerical integration and those that are directly formulated in 2D manifold form, like the Koiter, Canham and Helfrich models. Further, a unified approach to enforce the $G^1$-continuity between patches, fix the angle between surface folds, enforce symmetry conditions and prescribe rotational Dirichlet boundary conditions, is presented using penalty and Lagrange multiplier methods. The formulation is fully described in the natural curvilinear coordinate system of the finite element description, which facilitates an efficient computational implementation. It contains existing isogeometric thin shell formulations as special cases. Several classical numerical benchmark examples are considered to demonstrate the robustness and accuracy of the proposed formulation. The presented constitutive models, in particular the simple mixed Koiter model that does not require any thickness integration, show excellent performance, even for large deformations.

2501.04856

James Newland

James Newland (Texas Advanced Computing Center, University of Texas at Austin)

Using Data Science in High School Astronomy

Conference paper for session at ASP 2024 virtual conference

physics.ed-ph astro-ph.IM astro-ph.SR

http://creativecommons.org/licenses/by-nc-sa/4.0/

Astronomy datasets can be challenging to use for high school astronomy classes. Data science education pedagogy can be leveraged to create astronomy activities in which students interrogate data, create visuals, and use statistical thinking to construct astronomy knowledge. This session describes how the NASA/IPAC Infrared Science Archive (IRSA) can provide a web-based interface for students to use basic data science techniques in astronomy to build data literacy while learning astronomical concepts. The activities shared will be available for anyone but were designed to be used in astro 101 classes in high school or early college.

2501.04857

Antonin Eddi

S\'ebastien Kuchly, Wilson Reino, Kyle McKee, St\'ephane Perrard, Giuseppe Pucci and Antonin Eddi

Resonance of an object floating within a surface wavefield

physics.flu-dyn

http://creativecommons.org/licenses/by/4.0/

We examine the interaction between floating cylindrical objects and surface waves in the gravity regime. Given that the impact of resonance phenomena associated with floating bodies, particularly at laboratory scales, remains underexplored, we focus on the influence of the floaters' resonance frequency on wave emission. We study the response of floating rigid cylinders to external mechanical perturbations. Using an optical reconstruction technique to measure wave fields, we study the natural resonance frequency of the floaters in geometry and their response to incoming waves. The results indicate that the resonance frequency is influenced by the interplay between the cylinder geometry and the solid-to-fluid density ratio. Experiments with incoming waves demonstrate that floaters diffract waves, creating complex patterns of diffusion and retro-diffusion while radiating secondary waves that interfere with the incident wavefield. Minimal wave generation is observed at resonance frequencies. These findings provide insights for elucidating the behavior of larger structures, such as sea ice floes, in natural wave fields.

2501.04858

Mohammad Amin Abbasi

Sara Bourbour Hosseinbeigi, Sina Asghari, Mohammad Ali Seif Kashani, Mohammad Hossein Shalchian, Mohammad Amin Abbasi

Advancing Retrieval-Augmented Generation for Persian: Development of Language Models, Comprehensive Benchmarks, and Best Practices for Optimization

cs.CL

http://creativecommons.org/licenses/by-nc-sa/4.0/

This paper examines the specific obstacles of constructing Retrieval-Augmented Generation(RAG) systems in low-resource languages, with a focus on Persian's complicated morphology and versatile syntax. The research aims to improve retrieval and generation accuracy by introducing Persian-specific models, namely MatinaRoberta(a masked language model) and MatinaSRoberta(a fine-tuned Sentence-BERT), along with a comprehensive benchmarking framework. Three datasets-general knowledge(PQuad), scientifically specialized texts, and organizational reports, were used to assess these models after they were trained on a varied corpus of 73.11 billion Persian tokens. The methodology involved extensive pretraining, fine-tuning with tailored loss functions, and systematic evaluations using both traditional metrics and the Retrieval-Augmented Generation Assessment framework. The results show that MatinaSRoberta outperformed previous embeddings, achieving superior contextual relevance and retrieval accuracy across datasets. Temperature tweaking, chunk size modifications, and document summary indexing were explored to enhance RAG setups. Larger models like Llama-3.1 (70B) consistently demonstrated the highest generation accuracy, while smaller models faced challenges with domain-specific and formal contexts. The findings underscore the potential for developing RAG systems in Persian through customized embeddings and retrieval-generation settings and highlight the enhancement of NLP applications such as search engines and legal document analysis in low-resource languages.

2501.04859

Lars Rohwedder

Lars Rohwedder

ETH-Tight FPT Algorithm for Makespan Minimization on Uniform Machines

cs.DS

http://creativecommons.org/publicdomain/zero/1.0/

Given $n$ jobs with processing times $p_1,\dotsc,p_n\in\mathbb N$ and $m\le n$ machines with speeds $s_1,\dotsc,s_m\in\mathbb N$ our goal is to allocate the jobs to machines minimizing the makespan. We present an algorithm that solves the problem in time $p_{\max}^{O(d)} n^{O(1)}$, where $p_{\max}$ is the maximum processing time and $d\le p_{\max}$ is the number of distinct processing times. This is essentially the best possible due to a lower bound based on the exponential time hypothesis (ETH). Our result improves over prior works that had a quadratic term in $d$ in the exponent and answers an open question by Kouteck\'y and Zink. The algorithm is based on integer programming techniques combined with novel ideas based on modular arithmetic. They can also be implemented efficiently for the more compact high-multiplicity instance encoding.

2501.04860

Michael F. Xu

Michael F. Xu and Bilge Mutlu

Exploring the Use of Robots for Diary Studies

Proceedings of the 29th ACM/IEEE International Conference on Human Robot Interaction (HRI 2025)

cs.RO cs.HC

http://creativecommons.org/licenses/by-nc-sa/4.0/

As interest in studying in-the-wild human-robot interaction grows, there is a need for methods to collect data over time and in naturalistic or potentially private environments. HRI researchers have increasingly used the diary method for these studies, asking study participants to self-administer a structured data collection instrument, i.e., a diary, over a period of time. Although the diary method offers a unique window into settings that researchers may not have access to, they also lack the interactivity and probing that interview-based methods offer. In this paper, we explore a novel data collection method in which a robot plays the role of an interactive diary. We developed the Diary Robot system and performed in-home deployments for a week to evaluate the feasibility and effectiveness of this approach. Using traditional text-based and audio-based diaries as benchmarks, we found that robots are able to effectively elicit the intended information. We reflect on our findings, and describe scenarios where the utilization of robots in diary studies as a data collection instrument may be especially applicable.

2501.04861

Hafiz Mughees Ahmad Mr.

Hafiz Mughees Ahmad and Dario Morle and Afshin Rahimi

LayerMix: Enhanced Data Augmentation through Fractal Integration for Robust Deep Learning

cs.CV

http://creativecommons.org/licenses/by-nc-nd/4.0/

Deep learning models have demonstrated remarkable performance across various computer vision tasks, yet their vulnerability to distribution shifts remains a critical challenge. Despite sophisticated neural network architectures, existing models often struggle to maintain consistent performance when confronted with Out-of-Distribution (OOD) samples, including natural corruptions, adversarial perturbations, and anomalous patterns. We introduce LayerMix, an innovative data augmentation approach that systematically enhances model robustness through structured fractal-based image synthesis. By meticulously integrating structural complexity into training datasets, our method generates semantically consistent synthetic samples that significantly improve neural network generalization capabilities. Unlike traditional augmentation techniques that rely on random transformations, LayerMix employs a structured mixing pipeline that preserves original image semantics while introducing controlled variability. Extensive experiments across multiple benchmark datasets, including CIFAR-10, CIFAR-100, ImageNet-200, and ImageNet-1K demonstrate LayerMixs superior performance in classification accuracy and substantially enhances critical Machine Learning (ML) safety metrics, including resilience to natural image corruptions, robustness against adversarial attacks, improved model calibration and enhanced prediction consistency. LayerMix represents a significant advancement toward developing more reliable and adaptable artificial intelligence systems by addressing the fundamental challenges of deep learning generalization. The code is available at https://github.com/ahmadmughees/layermix.

2501.04862

Steffen Rulands Prof. Dr.

Felix Meige, Lina Hellwig, Harald Stachelscheid, Philipp Mergenthaler, Steffen Rulands

Universal quasi-particle kinetics control the cell death decision

physics.bio-ph cond-mat.stat-mech q-bio.CB

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Understanding how fluctuations propagate across spatial scales is central to our understanding of inanimate matter from turbulence to critical phenomena. In contrast to physical systems, biological systems are organized into a hierarchy of processes on a discrete set of spatial scales: they are compartmentalized. Here, we show that dynamic compartmentalization of stochastic systems leads to emergent, quasi-particle-like kinetics which are used by cells to perform key biological functions. Specifically, we derive a general theory that predicts the emergence of a single degree of freedom irrespective of system specifics. We obtain equations of motion and response characterising its unique kinetic properties. We experimentally demonstrate the biological relevance of quasi-particle kinetics in the decision of cells to commit suicide (apoptosis). Using fluorescent microscopy, we show that the response of cells to apoptotic stimuli exhibits quasi-particle like kinetics which establish a low-pass filter for cellular stress signals. By highlighting that cells manipulate how noise and signals propagate across spatial scales, our work reveals a new mechanism of cell fate decision-making.

2501.04863

Rafayel Teymurazyan

Dami\~ao J. Ara\'ujo and Rafayel Teymurazyan

Interacting free boundaries in obstacle problems

19

math.AP

http://creativecommons.org/licenses/by/4.0/

We study obstacle problems governed by two distinct types of diffusion operators involving interacting free boundaries. We obtain a somewhat surprising coupling property, leading to a comprehensive analysis of the free boundary. More precisely, we show that near regular points of a coordinate function, the free boundary is analytic, whereas singular points lie on a smooth manifold. Additionally, we prove that uncoupled free boundary points are singular, indicating that regular points lie exclusively on the coupled free boundary. Furthermore, optimal regularity, non-degeneracy, and lower dimensional Hausdorff measure estimates are obtained. Explicit examples illustrate the sharpness of assumptions.

2501.04864

Matteo Giacomini

Matteo Giacomini, Davide Cortellessa, Luan M. Vieira, Ruben Sevilla, Antonio Huerta

A hybrid pressure formulation of the face-centred finite volume method for viscous laminar incompressible flows

44 pages, 18 figures, 6 tables

math.NA cs.CE cs.NA physics.flu-dyn

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This work presents a hybrid pressure face-centred finite volume (FCFV) solver to simulate steady-state incompressible Navier-Stokes flows. The method leverages the robustness, in the incompressible limit, of the hybridisable discontinuous Galerkin paradigm for compressible and weakly compressible flows to derive the formulation of a novel, low-order face-based discretisation. The incompressibility constraint is enforced in a weak sense, by introducing an inter-cell mass flux defined in terms of a new, hybrid variable, representing the pressure at the cell faces. This results in a new hybridisation strategy where cell variables (velocity, pressure and deviatoric strain rate tensor) are expressed as a function of velocity and pressure at the barycentre of the cell faces. The hybrid pressure formulation provides first-order convergence of all variables, including the stress, independently of cell type, stretching and distortion. Numerical benchmarks of Navier-Stokes flows at low and moderate Reynolds numbers, in two and three dimensions, are presented to evaluate accuracy and robustness of the method. In particular, the hybrid pressure formulation outperforms the FCFV method when convective effects are relevant, achieving accurate predictions on significantly coarser meshes.

2501.04865

Alexander Boyd

Alexander B. Boyd and Paul M. Riechers

Time Symmetries of Quantum Memory Improve Thermodynamic Efficiency

cond-mat.stat-mech quant-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Classical computations inherently require energy dissipation that increases significantly as the reliability of the computation improves. This dissipation arises when transitions between memory states are not balanced by their time-reversed counterparts. While classical memories exhibit a discrete set of possible time-reversal symmetries, quantum memory offers a continuum. This continuum enables the design of quantum memories that minimize irreversibility. As a result, quantum memory reduces energy dissipation several orders of magnitude below classical memory.

2501.04866

Tayt Armitage

Tayt Armitage, David V. Martin, Romy Rodr\'iguez Mart\'inez

Identifying Flare Locations Through Exoplanet Transit Occultations

Under review at MNRAS updated in regards to third referee report. 16 pages. 12 figures

astro-ph.EP astro-ph.SR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

M dwarfs are the most common stars in the galaxy, with long lifespans, a high occurrence rate of rocky planets, and close-in habitable zones. However, high stellar activity in the form of frequent flaring and any associated coronal mass ejections may drive atmospheric escape with the bombardment of radiation and high-energy particles, drastically impacting the habitability of these systems. The stellar latitude where flares and coronal mass ejections occur determines the space weather that exoplanets are subject to, with high-energy particle events associated with equatorial flares producing significant atmospheric erosion. However, the flaring latitudes for M dwarfs remain largely unconstrained. To aid in the effort to locate these flaring regions we explore the applicability of flare occultations using optical photometry to identify the latitudes of flares. As a planet transits in front of an ongoing flare the timing and geometry of the transit can be used to constrain the latitude and longitude of the flare. We predict the probability of detecting an occultation for known transiting planets and eclipsing binaries. From this, we estimate 3-22 detectable occultations exist within the TESS primary mission photometry, with the majority occurring in eclipsing binary observations. To demonstrate this technique, we analyze a candidate flare occultation event for the eclipsing binary CM Draconis.

2501.04867

Serge Tabachnikov

Serge Tabachnikov

On cusps of caustics by reflection in two dimensional projective Finsler metrics

math.DG math.DS

http://creativecommons.org/licenses/by/4.0/

A Finsler, not necessarily symmetric, metric in the plane or its convex subset is called projective if its geodesics are straight segments. We consider Finsler billiards in a convex planar domain endowed with a projective Finsler metric. A caustic by reflection is the envelope of the oriented lines, the billiard trajectories, that start at a point inside the billiard and undergo a fixed number of reflections. We show that such a caustic has at least four cusps. This problem is motivated by the "Last Geometric Statement of Jacobi" that the conjugate locus of a non-umbilic point of a triaxial ellipsoid has exactly four cusps. The present note extends the recent results in this direction concerning Euclidean billiards.

2501.04868

Pete Riley

Pete Riley, Roberto Lionello, and Yemmy Rivera

Ion Charge States from a Global Time-Dependent Wave-Turbulence-Driven Model of the Solar Wind: Comparison with in-situ Measurements

physics.space-ph

http://creativecommons.org/licenses/by/4.0/

Solar wind charge-state measurements contain a wealth of knowledge related to the properties of the solar corona from where they originated. However, their interpretation has remained challenging because it convolves coronal temperature, density, and velocity along the particles' trajectory through the corona before they ``freeze in'' and are convected outward through the solar wind. In this study, we calculate ion charge states by coupling a non-equilibrium ionization model with a global magnetohydrodynamic model of the corona and inner heliosphere. We present results for two periods characteristic of solar minimum and maximum and compare them with observations from the ACE spacecraft. We find that the model reproduces the essential features of the observations, rectifying an earlier inconsistency that was apparent in 1-D calculations, and allows us to unambiguously trace the evolution of charge states from the base of the corona into the solar wind.

2501.04869

Olga Zolotareva

Olga Zolotareva, Karen Legler, Olga Tsoy, Anna Esteve, Alexey Sergushichev, Vladimir Sukhov, Jan Baumbach, Kathrin Eylmann, Minyue Qi, Malik Alawi, Stefan Kommoss, Barbara Schmalfeldt, Leticia Oliveira-Ferrer

Transcriptome signature for the identification of bevacizumab responders in ovarian cancer

q-bio.GN

http://creativecommons.org/licenses/by/4.0/

The standard of care for ovarian cancer comprises cytoreductive surgery, followed by adjuvant platinum-based chemotherapy plus taxane therapy and maintenance therapy with the antiangiogenic compound bevacizumab and/or a PARP inhibitor. Nevertheless, there is currently no clear clinical indication for the use of bevacizumab, highlighting the urgent need for biomarkers to assess the response to bevacizumab. In the present study, based on a novel RNA-seq dataset (n=181) and a previously published microarray-based dataset (n=377), we have identified an expression signature potentially associated with benefit from bevacizumab addition and assumed to reflect cancer stemness acquisition driven by activation of CTCFL. Patients with this signature demonstrated improved overall survival when bevacizumab was added to standard chemotherapy in both novel (HR=0.41(0.23-0.74), adj.p-value=7.70e-03) and previously published cohorts (HR=0.51(0.34-0.75), adj.p-value=3.25e-03), while no significant differences in survival explained by treatment were observed in patients negative for this signature. In addition to the CTCFL signature, we found several other reproducible expression signatures which may also represent biomarker candidates not related to established molecular subtypes of ovarian cancer and require further validation studies based on additional RNA-seq data.

2501.04870

Elynn Chen

Jinhang Chai, Elynn Chen, Jianqing Fan

Deep Transfer $Q$-Learning for Offline Non-Stationary Reinforcement Learning

stat.ML cs.LG

http://creativecommons.org/licenses/by/4.0/

In dynamic decision-making scenarios across business and healthcare, leveraging sample trajectories from diverse populations can significantly enhance reinforcement learning (RL) performance for specific target populations, especially when sample sizes are limited. While existing transfer learning methods primarily focus on linear regression settings, they lack direct applicability to reinforcement learning algorithms. This paper pioneers the study of transfer learning for dynamic decision scenarios modeled by non-stationary finite-horizon Markov decision processes, utilizing neural networks as powerful function approximators and backward inductive learning. We demonstrate that naive sample pooling strategies, effective in regression settings, fail in Markov decision processes.To address this challenge, we introduce a novel ``re-weighted targeting procedure'' to construct ``transferable RL samples'' and propose ``transfer deep $Q^*$-learning'', enabling neural network approximation with theoretical guarantees. We assume that the reward functions are transferable and deal with both situations in which the transition densities are transferable or nontransferable. Our analytical techniques for transfer learning in neural network approximation and transition density transfers have broader implications, extending to supervised transfer learning with neural networks and domain shift scenarios. Empirical experiments on both synthetic and real datasets corroborate the advantages of our method, showcasing its potential for improving decision-making through strategically constructing transferable RL samples in non-stationary reinforcement learning contexts.

2501.04871

Kaitlyn Lee

Kaitlyn J. Lee and Alejandro Schuler

RieszBoost: Gradient Boosting for Riesz Regression

stat.ML cs.LG stat.ME

http://creativecommons.org/licenses/by/4.0/

Answering causal questions often involves estimating linear functionals of conditional expectations, such as the average treatment effect or the effect of a longitudinal modified treatment policy. By the Riesz representation theorem, these functionals can be expressed as the expected product of the conditional expectation of the outcome and the Riesz representer, a key component in doubly robust estimation methods. Traditionally, the Riesz representer is estimated indirectly by deriving its explicit analytical form, estimating its components, and substituting these estimates into the known form (e.g., the inverse propensity score). However, deriving or estimating the analytical form can be challenging, and substitution methods are often sensitive to practical positivity violations, leading to higher variance and wider confidence intervals. In this paper, we propose a novel gradient boosting algorithm to directly estimate the Riesz representer without requiring its explicit analytical form. This method is particularly suited for tabular data, offering a flexible, nonparametric, and computationally efficient alternative to existing methods for Riesz regression. Through simulation studies, we demonstrate that our algorithm performs on par with or better than indirect estimation techniques across a range of functionals, providing a user-friendly and robust solution for estimating causal quantities.

2501.04872

Oleg Titov

Oleg Titov and Angelina Osetrova

Parallactic delay for geodetic VLBI and non-orthogonality of the fundamental axes

Published in Publications of the Astronomical Society of Australia

Publications of the Astronomical Society of Australia, Volume 41, 2024, e111

10.1017/pasa.2024.111

astro-ph.IM astro-ph.GA

http://creativecommons.org/licenses/by/4.0/

The Gaia optical astrometric mission has measured the precise positions of millions of objects in the sky, including extragalactic sources also observed by Very Long Baseline Interferometry (VLBI). In the recent Gaia EDR3 release, an effect of negative parallax with a magnitude of approximately -17 $\mu$as was reported, presumably due to technical reasons related to the relativistic delay model. A recent analysis of a 30-year set of geodetic VLBI data revealed a similar negative parallax with an amplitude of $-15.8 \pm 0.5$ $ \mu$as. Since both astrometric techniques, optical and radio, provide consistent estimates of this negative parallax, it is necessary to investigate the potential origin of this effect. We developed the extended group relativistic delay model to incorporate the additional parallactic effect for radio sources at distances less than 1 Mpc and found that the apparent annual signal might appear due the non-orthogonality of the fundamental axes, which are defined by the positions of the reference radio sources themselves. Unlike the conventional parallactic ellipse, the apparent annual effect in this case appears as a circular motion for all objects independently of their ecliptic latitude. The measured amplitude of this circular effect is within a range of 10-15 $\mu$as that is consistent with the ICRF3 stability of the fundamental axis. This annual circular effect could also arise if a G\"odel-type cosmological metric were applied, suggesting that, in the future, this phenomenon could be used to indicate global cosmic rotation.

2501.04873

Alexander Gabriel Valverde Guillen

Alexander Valverde and Luis Solano

Back Home: A Machine Learning Approach to Seashell Classification and Ecosystem Restoration

cs.CV cs.AI cs.LG

http://creativecommons.org/licenses/by-nc-nd/4.0/

In Costa Rica, an average of 5 tons of seashells are extracted from ecosystems annually. Confiscated seashells, cannot be returned to their ecosystems due to the lack of origin recognition. To address this issue, we developed a convolutional neural network (CNN) specifically for seashell identification. We built a dataset from scratch, consisting of approximately 19000 images from the Pacific and Caribbean coasts. Using this dataset, the model achieved a classification accuracy exceeding 85%. The model has been integrated into a user-friendly application, which has classified over 36,000 seashells to date, delivering real-time results within 3 seconds per image. To further enhance the system's accuracy, an anomaly detection mechanism was incorporated to filter out irrelevant or anomalous inputs, ensuring only valid seashell images are processed.