October 2025 arXiv papers — page 40

Wai Yeung Lam, Masashi Yasumoto

We survey structure-preserving discretizations of minimal surfaces in Euclidean space. Our focus is on a discretization defined via parallel face offsets of polyhedral surfaces, which naturally leads to a notion of vanishing mean curvature and a corresponding variational characterization. All simply connected discrete minimal surfaces of this type can be con

Nicki Barari, Edward Kim, Christopher MacLellan

Catastrophic forgetting remains a central challenge in continual learning, where models are required to integrate new knowledge over time without losing what they have previously learned. In prior work, we introduced Cobweb/4V, a hierarchical concept formation model that exhibited robustness to catastrophic forgetting in visual domains. Motivated by this rob

Piyali Chakraborty, Dorin Ervin Dutkay

In relation to Fuglede's conjecture, we establish several Plancherel-type identities and demonstrate the surjectivity of the Fourier transform between certain unbounded tiling sets of $\mathbb{R}$ that are in duality. In the terminology commonly used in the context of Fuglede's conjecture, our result states that an open set tiles $\mathbb{R}$ by the finite s

Judy Shir, Shahar Hadar

The unstably bound, critical null geodesics of the Kerr spacetime form a distinguished class of orbits whose properties govern observables such as the photon ring and the high-frequency component of black-hole ringdown. This set of orbits defines a codimension-two submanifold of the null-geodesic phase space known as the photon shell. In this work we investi

Friedrich Martin Schneider

We prove that the projective unit group $\mathrm{PGL}(R)$, i.e., the quotient of the unit group $\mathrm{GL}(R)$ modulo its center, of any non-discrete irreducible, continuous ring $R$ is simple. Moreover, we show that $\mathrm{GL}(R)$ has uncountable strong cofinality, that is, it is not the union of a countable chain of proper subgroups and it has finite w

Ignavier Ng, Patrick Blöbaum, Siddharth Bhandari, Kun Zhang

Recent alignment techniques, such as reinforcement learning from human feedback, have been widely adopted to align large language models with human preferences by learning and leveraging reward models. In practice, these models often exploit spurious correlations, involving, e.g., response length, discrimination, sycophancy, and conceptual bias, which is a p

Gagan Bansal, Wenyue Hua, Zezhou Huang, Adam Fourney

As LLM agents advance, they are increasingly mediating economic decisions, ranging from product discovery to transactions, on behalf of users. Such applications promise benefits but also raise many questions about agent accountability and value for users. Addressing these questions requires understanding how agents behave in realistic market conditions. Howe

Chunran Zhu, Bolin Li

The Equation of State (EOS) of matter within neutron stars is a central topic in nuclear physics and astrophysics.This study investigates hadron-quark hybrid stars by integrating the density-dependent DDME2 relativistic mean-field model for hadronic matter with a two-flavor Nambu-Jona-Lasinio (NJL) model for quark matter.A quintic polynomial interpolation is

John F. Wu, Michael Walmsley

Sparse Autoencoders (SAEs) can efficiently identify candidate monosemantic features from pretrained neural networks for galaxy morphology. We demonstrate this on Euclid Q1 images using both supervised (Zoobot) and new self-supervised (MAE) models. Our publicly released MAE achieves superhuman image reconstruction performance. While a Principal Component Anal

Jose R. Nascimento, Ana R. M. Oliveira, Albert Yu. Petrov, Paulo J. Porfírio

The existence of black hole shadows is one of the most interesting effects of the strong field regime of general relativity (GR). Recent observations by the Event Horizon Telescope (EHT) have provided high-resolution images of the vicinity of supermassive black holes, ushering in a new era for testing gravitation on astrophysical scales. In this work, we con

A. Lemière, G. Castelletti, N. L. Maza

We present the first detection of GeV $\gamma$-ray emission potentially associated with the pulsar wind nebula (PWN) hosted by the young core-collapse supernova remnant G292.0+1.8, based on a detailed time-resolved analysis of \textit{Fermi}-LAT data. By isolating the unpulsed component from the dominant magnetospheric radiation of PSR~J1124$-$5916, we succe

William Oswald, Mario Renteria-Pinon, Md. Sajjad Hossain, Kyle Mooney

Bit truncation has demonstrated great potential to enable run-time quality-power adaptive data storage, thereby optimizing the power/energy efficiency of approximate applications and supporting their deployment in edge environments. However, existing bit-truncation memories require custom designs for a specific application. In this paper, we present a novel

Laura Mismetti, Marvin Alberts, Andreas Krause, Mara Graziani

Tandem Mass Spectrometry is a cornerstone technique for identifying unknown small molecules in fields such as metabolomics, natural product discovery and environmental analysis. However, certain aspects, such as the probabilistic fragmentation process and size of the chemical space, make structure elucidation from such spectra highly challenging, particularl

Alex Alberts, Ilias Bilionis

Quantifying the uncertainty in the output of a neural network is essential for deployment in scientific or engineering applications where decisions must be made under limited or noisy data. Bayesian neural networks (BNNs) provide a framework for this purpose by constructing a Bayesian posterior distribution over the network parameters. However, the prior, wh

Eline M. Bovy, Caleb Probine, Marnix Suilen, Ufuk Topcu

Multi-environment POMDPs (ME-POMDPs) extend standard POMDPs with discrete model uncertainty. ME-POMDPs represent a finite set of POMDPs that share the same state, action, and observation spaces, but may arbitrarily vary in their transition, observation, and reward models. Such models arise, for instance, when multiple domain experts disagree on how to model

Makoto A. Johnstone, Ezequiel Treister, Franz E. Bauer, Chin-Shin Chang

We present high-resolution ($\sim$50$-$100 pc) Atacama Large Millimeter Array (ALMA) observations of $^{12}$CO(2-1) or $^{12}$CO(1-0) emission in seven local ($z$ $\lesssim$ 0.05) major mergers -- five of which are dual active galactic nuclei (AGN) systems, and two of which are single AGN systems. We model the molecular gas kinematics through rotating disk p

Jun-Sung Moon, Teppei Okumura

The tidal torque theory (TTT) predicts that galaxy spins are correlated with the surrounding tidal field, reflecting how angular momentum is acquired during structure formation. We present a new observational test of this prediction using the final data release of the Sloan Digital Sky Survey IV Mapping Nearby Galaxies at Apache Point Observatory integral fi

HaoYang Shang, Xuan Liu

The rapid evolution of large language models (LLMs) creates complex bidirectional expectations between users and AI systems that are poorly understood. We introduce the concept of "mutual wanting" to analyze these expectations during major model transitions. Through analysis of user comments from major AI forums and controlled experiments across multiple Ope

Shay Gilpin, Michael Herty

Particle filters are a widely used Monte Carlo based data assimilation technique that estimates the probability distribution of a system's state conditioned on observations through a collection of weights and particles. A known problem for particle filters is weight collapse, or degeneracy, where a single weight attains a value of one while all others are cl

Yan Liu, Jonathan Chuah, Michael Unser, Jonathan Dong

In this paper, we propose a global framework that includes a detailed model of the photo-switching and acoustic processes for photo-switching optoacoustic mesoscopy, based on the underlying physics. We efficiently implement two forward models as matrix-free linear operators and join them as one forward operator. Then, we reconstruct the concentration maps di

Camilla Mupo, Yue Zhang

The QCD axion can successfully solve the strong CP problem under the condition that the Peccei-Quinn symmetry is respected to extremely high standard. We explore a class of gauge theories that accommodate a high-quality axion, known as the Barr-Seckel models, paying special attention to the cosmology of topological defects. In models with domain wall number

Fuat Can Beylunioglu, P. Robert Duimering, Mehrdad Pirnia

Deep neural networks (DNNs) have been used to model complex optimization problems in many applications, yet have difficulty guaranteeing solution optimality and feasibility, despite training on large datasets. Training a NN as a surrogate optimization solver amounts to estimating a global solution function that maps varying problem input parameters to the co

Arthur Ji Sung Baek, Geoffrey Martin

We present X-SYCON, a xylem-inspired multi-agent architecture in which coordination emerges from passive field dynamics rather than explicit planning or communication. Incidents (demands) and obstructions (hazards) continually write diffusing and decaying scalar fields, and agents greedily ascend a local utility $U=\phi_{\mathrm{DE}}-\kappa\,\phi_{\mathrm{HZ

Stephane Dartois, Gilles Zémor

In this paper, we compute the injective norm - a.k.a. geometric entanglement - of standard basis states of CSS quantum error-correcting codes. The injective norm of a quantum state is a measure of genuine multipartite entanglement. Computing this measure is generically NP-hard. However, it has been computed exactly in condensed-matter theory - notably in the

Jasper M. Liu, Hai Zhu

For fixed positive integers $n,m,r$, let $\mathrm{Mat}_{n \times m}(\mathbb{C})$ be the affine space of $n \times m$ complex matrices with coordinate ring $\mathbb{C}[\mathbf{x}_{n \times m}]$. We define a homogeneous ideal $I_{n,m,r}$, where the graded quotient $\mathbb{C}[\mathbf{x}_{n \times m}]/I_{n,m,r}$ is obtained from the orbit harmonics deformation

Eamon Duede

This paper examines the role of artificial intelligence in scientific problem-solving, with a focus on its implications for disciplinary creativity. Drawing on recent work in the philosophy of creativity, I distinguish between creative approaches and creative products, and introduce the concept of disciplinary creativity -the creative application of discipli

David R. Rice, Chenliang Huang, Robert Royer, Mangesh Daspute

Magrathea is an open-source C++ code for modeling the internal structure of differentiated planets. The initial release, Huang et al. (2022), introduced the base solver, a modular framework for defining equations of state (EOSs) used within phase diagrams for each differentiated layer, and outlined a series of planned extensions. Many of those updates are no

Mansi M. Kasliwal, Tomas Ahumada, Robert Stein, Viraj Karambelkar

On August 18, 2025, the LIGO-Virgo-KAGRA collaboration reported gravitational waves from a sub-threshold binary neutron star merger. If astrophysical, this event would have a surprisingly low chirp mass, suggesting that at least one neutron star was below a solar mass. The Zwicky Transient Facility mapped the coarse localization and discovered a transient, Z

Himanshu Badhani, Dhanuja G S, Siddhartha Das

We introduce the resource-theoretic free energy of a quantum channel as the maximal work extractable from the channel as its output equilibrates to a thermal state and its reference system remains locally intact. It is proportional to the relative entropy between the given channel and the absolutely thermal channel. It attains a clear operational meaning as

Alessandra Terranova, Björn Ross, Alexandra Birch

In order for large language models to achieve true conversational continuity and benefit from experiential learning, they need memory. While research has focused on the development of complex memory systems, it remains unclear which types of memory are most effective for long-context conversational tasks. We present a systematic evaluation of memory-augmente

Ian P. A. Johnson, Taeho Ryu, Rosalba Perna

We present hydrodynamic simulations with the moving-mesh code AREPO of Tidal Disruption Encores (TDEEs) in nuclear star clusters (NSCs). TDEEs arise when a stellar-mass black hole (sBH) disrupts a star within the NSC, producing debris that is unbound from the sBH but remains gravitationally bound to the central massive black hole (MBH), leading to a delayed

Brendan O'Connor, Roberto Ricci, Eleonora Troja, Antonella Palmese

The first joint electromagentic (EM) and gravitational wave (GW) detection, known as GW170817, marked a critical juncture in our collective understanding of compact object mergers. However, it has now been 8 years since this discovery, and the search for a second EM-GW detection has yielded no robust discoveries. Recently, on August 18, 2025, the LIGO-Virgo-

Anisha Saha, Varsha Suresh, Timothy Hospedales, Vera Demberg

Sarcasm is a specific type of irony which involves discerning what is said from what is meant. Detecting sarcasm depends not only on the literal content of an utterance but also on non-verbal cues such as speaker's tonality, facial expressions and conversational context. However, current multimodal models struggle with complex tasks like sarcasm detection, w

Bilal Hawashin, Max Uetrecht

It was recently established that the paradigmatic Gross--Neveu model with $N$ copies of two-dimensional Dirac fermions features an $\mathrm{SO}(2N)$ symmetry if certain interactions are suppressed. This becomes evident when the theory is rewritten in terms of $2N$ copies of two-dimensional Majorana fermions. Mean-field theory for the $\mathrm{SO}(2N)$ model

Mayusree Das, Banibrata Mukhopadhyay, Tomasz Bulik

White dwarfs (WDs) are frequently observed to have strong magnetic fields up to $10^9$ G and expected to have a possible internal field as high as $\sim 10^{14}$ G. High internal fields can significantly deform a WD's equilibrium structure, generating a quadrupole moment. If the rotation axis is misaligned with the magnetic axis, the deformation can lead to

Xander J. Hall, Antonella Palmese, Brendan O'Connor, Daniel Gruen

On August 18th, 2025, the LIGO--Virgo--KAGRA collaboration reported a sub-threshold gravitational wave candidate detection consistent with a sub-solar-mass neutron star merger, denoted S250818k. An optical transient, AT2025ulz, was discovered within the localization region. AT2025ulz initially appeared to meet the expected behavior of kilonova (KN) emission,

Kaito Iwasaki, Anthony Bloch, Maani Ghaffari

This paper introduces a convex optimization framework for identifying switched network systems, in which both the node dynamics and the underlying graph topology switch between a finite number of configurations. Building on our recent convex identification method for general switching systems, we extend the formulation to structured network systems where eac

Xiang Li, Ting-Chun Lin, Yahya Alavirad, John McGreevy

We propose an operator generalization of the Li-Haldane conjecture regarding the entanglement Hamiltonian of a disk in a 2+1D chiral gapped groundstate. The logic applies to regions with sharp corners, from which we derive several universal properties regarding corner entanglement. These universal properties follow from a set of locally-checkable conditions

Aiden Sheckler

We point out that there exists a generalization of instanton symmmetry in the Coulomb phase of 5d nonabelian gauge theories which is capable of measuring a wider class of topological charges of monopole strings. The symmetry is invertible on compact spacetimes, but non-invertible on spacetimes with boundary. In the case of maximal supersymmetry, we show that

Yeonju Go, Dmitrii Torbunov, Yi Huang, Shuhang Li

Accurate separation of signal from background is one of the main challenges for precision measurements across high-energy and nuclear physics. Conventional supervised learning methods are insufficient here because the required paired signal and background examples are impossible to acquire in real experiments. Here, we introduce an unsupervised unpaired imag

Klée Pollock, Jonathan D. Kroth, Jonathon Riddell, Thomas Iadecola

We study one dimensional quantum spin chains whose nearest neighbor interactions are random matrices that square to one. By employing free probability theory, we establish a mapping from the many-body quantum dynamics of energy density in the original chain to a single-particle hopping dynamics when the local Hilbert space dimension is large. The hopping occ

Joaquin Masias

We study the scalar potentials that arise from higher curvature corrections in general $f(R)$ theories of gravity and their connection to a dynamical species scale. Starting from general considerations in arbitrary dimensions, we show that at large field values, the scalar potential generated by an infinite series of curvature terms and the field dependent s

Saiyad Ashanujjaman, Guglielmo Coloretti, Andreas Crivellin, Siddharth P. Maharathy

The decay $A\to ZH$ is a characteristic signal of two-Higgs-doublet models (2HDMs), where $A$ and $H$ lie primarily within the same $SU(2)_L$ multiplet, leading to a coupling of order $g_2$ to the $Z$ boson. The subsequent decay $H\to tt^{(*)}$ is particularly promising, as it gives rise to distinct final states involving multiple leptons and $b$-jets. The r

Hooman Davoudiasl, Peter B. Denton, Anna M. Suliga

The growing evidence for nano-hertz gravitational waves, from NANOGrav and other observations, may be pointing to a cosmological first-order phase transition at temperatures of $\mathcal{O}(10-100)\;\mathrm{MeV}$. Such an interpretation requires dynamics beyond the Standard Model in this energy range. If so, it may well be the case that core-collapse superno

Rishav Roshan, Sudipta Show

While most studies of the \textit{minimal type-I seesaw} neglect the heaviest decoupled right-handed neutrino, assuming negligible contributions to neutrino masses and leptogenesis, we demonstrate that its cosmological role can be significant. When long-lived, the presence of this particle can substantially modify the charged lepton Yukawa equilibration temp

Ana Bokulić, Tajron Jurić, Ivica Smolić

The search for regular black holes with nonlinear electromagnetic fields has sprouted numerous candidates, each exhibiting certain virtues but often accompanied by significant drawbacks. We demonstrate that Komar mass, electric charge and magnetic charge are mutually dependent in regular black holes with nonlinear electromagnetic fields, defined by a Lagrang

Joshuah T. Heath, Alexander C. Tyner, S. Pamir Alpay, Peter Krogstrup

We investigate the impact of two-level systems (TLSs) on superconductivity, treating them as soft modes localised in real space. We show that these defects can either enhance or suppress the superconducting critical temperature, depending on their surface density and average frequency. Using thin-film aluminium as a case study, we quantitatively describe how

Andrew Householder, Mary Anne Limbach, Beth Biller, Brooke Kotten

JWST's unprecedented sensitivity enables precise spectrophotometric monitoring of substellar worlds, revealing atmospheric variability driven by mechanisms operating across different pressure levels. This same precision now permits exceptionally sensitive searches for transiting exosatellites, small terrestrial companions to these worlds. Using a novel simul

Ritvik Sai Narayan, Melinda Soares-Furtado, Mary Anne Limbach, Nishanth Ramanujam

Young stars host only a small fraction of the known exoplanet population because their photometric variability, magnetic activity, and frequent placement in dense, poorly-resolved regions hamper exoplanet detections. Yet, measuring planets at these ages is crucial since these phases are when dynamical processes that drive planetary migration are most active.

Mubela Mutale, Mark A. Thompson, Gwenllian M. Williams, Andrew J. Rigby

We present a catalogue of compact sources detected in the SARAO MeerKAT 1.3 GHz Galactic Plane Survey (SMGPS). We extract 510599 compact sources, with areas less than five 8" beams, from the survey maps covering the regions $252\deg< l < 358^\deg$ and $2\deg < l < 61\deg$ at $|b| \leq 1.5^\deg$, which have an angular resolution of 8" and a sensitivity of $\s

Marc Kamionkowski, Kris Sigurdson

Self-gravitating galactic halos composed of self-interacting dark matter exhibit the formation of a highly dense core at the galactic center--a gravothermal collapse. Analytic models to describe this evolution have been developed and calibrated to numerical simulations initialized with isotropic particle velocity distributions, an assumption not necessarily

Niccolò Francini, Lukas Schmidt, Lukas Janssen, Daniel Lozano-Gómez

Pyrochlore magnets are a paradigmatic example of three-dimensional frustrated systems and provide an excellent platform for studying a variety of exotic many-body phenomena, including spin liquids, nematic phases, fragmentation, and order by disorder. In recent years, increasing attention has been devoted to bilinear spin models on this lattice, where multip

Gino Isidori, Paride Paradisi, Andrea Sainaghi, Nudzeim Selimovic

We investigate the neutrino sector in the framework of flavor deconstruction with an inverse-seesaw realization. This setup naturally links the hierarchical charged-fermion masses to the anarchic pattern of light-neutrino mixing. We determine the viable parameter space consistent with oscillation data and study the phenomenology of heavy neutral leptons (HNL

Siddharth Chaini, Federica B. Bianco, Ashish Mahabal

Distance-based methods involve the computation of distance values between features and are a well-established paradigm in machine learning. In anomaly detection, anomalies are identified by their large distance from normal data points. However, the performance of these methods often hinges on a single, user-selected distance metric (e.g., Euclidean), which m

Rebecca Diesing, Ludwig Böss, Damiano Caprioli

Merging galaxy clusters exhibit strong observational evidence for efficient particle acceleration in the intracluster medium (ICM), particularly in the form of synchrotron-emitting radio relics and halos. Cosmic ray (CR) electrons are likely accelerated (or re-accelerated) at merger and accretion shocks via diffusive shock acceleration (DSA). However, in the

Jan Boruch, Roberto Emparan, Luca V. Iliesiu, Sameer Murthy

We construct a series of novel Euclidean multi-black-hole, black ring, black Saturn, and black lens solutions to $5d$ supergravity that contribute as saddle-points to the $5d$ gravitational supersymmetric index, either in asymptotically flat space or in asymptotically AdS$_3\times S^2$. All these solutions are supersymmetric, have finite temperature, and an

David Berenstein, Joan Simon

The flat space limit of scalar bulk fields in AdS is discussed within a Lorentzian canonical quantization setup tailored to describe AdS state preparation and to extract the flat S-matrix dynamics. We discuss how the algebraic \`{I}n\"on\"u-Wigner contraction captures the local physics of the equivalence principle in quantum field theory in a fixed backgroun

Tae Hyun Jung, Takemichi Okui, Kohsaku Tobioka, Jiabao Wang

We study Big Bang Nucleosynthesis (BBN) constraints on heavy QCD axions. BBN offers a powerful probe of new physics that modifies the neutron-to-proton ratio during the process, thanks to the precisely measured primordial Helium-4 abundance. A heavy QCD axion provides an attractive target for this probe, because not only is it a well-motivated hypothetical p

Luigi "Rolly'' BEDIN

The nominal duration of Euclid's main mission is six years, but current best estimates indicate that the observatory has sufficient propellant to operate for up to ~14 years in total. In this work, we advocate dedicating six of these ~8 additional years to repeating the main survey, covering approximately one-third of the sky. This repetition would not only

Yujia Zhang, Xiaoyang Wu, Yixing Lao, Chengyao Wang

Humans learn abstract concepts through multisensory synergy, and once formed, such representations can often be recalled from a single modality. Inspired by this principle, we introduce Concerto, a minimalist simulation of human concept learning for spatial cognition, combining 3D intra-modal self-distillation with 2D-3D cross-modal joint embedding. Despite

Yichi Zhang, Alex Schwing, Zhizhen Zhao

Masked diffusion models have recently emerged as a flexible framework for discrete generative modeling. However, a key limitation of standard masked diffusion is its inability to effectively capture dependencies among tokens that are predicted concurrently, leading to degraded generation quality when dependencies among tokens are important. To explicitly mod

Shuhong Zheng, Ashkan Mirzaei, Igor Gilitschenski

Current 3D/4D generation methods are usually optimized for photorealism, efficiency, and aesthetics. However, they often fail to preserve the semantic identity of the subject across different viewpoints. Adapting generation methods with one or few images of a specific subject (also known as Personalization or Subject-driven generation) allows generating visu

Joachim Baumann

This PhD thesis investigates the societal impact of machine learning (ML). ML increasingly informs consequential decisions and recommendations, significantly affecting many aspects of our lives. As these data-driven systems are often developed without explicit fairness considerations, they carry the risk of discriminatory effects. The contributions in this t

Yi-Jie Wang, Geng-Dong Zhou, Hyunsung Jung, Seongyeon Youn

Recent experiments support that the magic-angle graphene can be modeled by a periodic array of correlated quantum impurities, immersed in a Dirac sea. This work analytically tackles a spin-valley Anderson impurity, featuring a general (anti-)Hund&#39;s interaction ($J_D, J_S$) that can originate from electron-phonon couplings. We derive its full phase diagra

Yuqian Yuan, Wenqiao Zhang, Xin Li, Shihao Wang

Multimodal large language models (MLLMs) have demonstrated strong general-purpose capabilities in open-world visual comprehension. However, most existing MLLMs primarily focus on holistic, scene-level understanding, often overlooking the need for fine-grained, object-centric reasoning. In this paper, we present PixelRefer, a unified region-level MLLM framewo

Calla Beauregard, Parisa Suchdev, Ashley M. A. Fehr, Isabelle T. Smith

Parental stress is a nationwide health crisis according to the U.S. Surgeon General's 2024 advisory. To allay stress, expecting parents seek advice and share experiences in a variety of venues, from in-person birth education classes and parenting groups to virtual communities, for example, BabyCenter, a moderated online forum community with over 4 million me

Ying-Hsuan Lin, Akira Tominaga, Mayuko Yamashita

We determine the $\mathrm{TMF}$-module structures of the genuine $C_2$-equivariant $\mathrm{TMF}$ with $\mathrm{RO}(C_2)$-gradings and of the $C_3$-equivariant $\mathrm{TMF}$. Moreover, we propose a general strategy for studying $C_n$-equivariant $\mathrm{TMF}$ via $U(1)$-equivariant $\mathrm{TMF}$ and a duality phenomenon in equivariant $\mathrm{TMF}$.

Jiahao Qiu, Xuan Qi, Hongru Wang, Xinzhe Juan

Large language models (LLMs) have been shown to perform better when scaffolded into agents with memory, tools, and feedback. Beyond this, self-evolving agents have emerged, but current work largely limits adaptation to prompt rewriting or failure retries. Therefore, we present ALITA-G, a self-evolution framework that transforms a general-purpose agent into a

En Chang, Vijay Pal Singh, Abel Beregi, Erik Rydow

The emergence of order in many-body systems and the associated self-similar dynamics governed by dynamical scaling laws is a hallmark of universality far from equilibrium. Measuring and classifying such nontrivial behavior for novel symmetry classes remains challenging. Here, we realize a well-controlled interlayer coupling quench in a tunable bilayer two-di

Hagen Papenburg

We construct local solutions to the Benjamin-Ono equation for quasi-periodic initial data. The solution is unique among limits of smooth solutions and depends continuously on the data. Our result applies to a richer class of quasi-periodic functions than previous theorems. Central to the argument is an a-priori estimate, the proof of which utilizes Strichart

Patrick R. Behr, Kevin France, Luca Fossati, Tommi Koskinen

KELT-20 b is an ultra-hot Jupiter with an equilibrium temperature of $2260$ K orbiting a bright (V =7.6), fast-rotating ($v\sin{i}$=117 km s$^{-1}$) A2 V star. The atmosphere of KELT-20 b has been studied extensively via transmission spectroscopy at optical wavelengths, showing strong hydrogen absorption as well as metals including Na I, Ca II, Fe I, Fe II,

Yizhu Jiao, Jiaqi Zeng, Julien Veron Vialard, Oleksii Kuchaiev

Large language models (LLMs) increasingly rely on thinking models that externalize intermediate steps and allocate extra test-time compute, with think-twice strategies showing that a deliberate second pass can elicit stronger reasoning. In contrast, most reward models (RMs) still compress many quality dimensions into a single scalar in one shot, a design tha

Yixing Chen, Yiding Wang, Siqi Zhu, Haofei Yu

Reinforcement Learning (RL) has demonstrated significant potential in enhancing the reasoning capabilities of large language models (LLMs). However, the success of RL for LLMs heavily relies on human-curated datasets and verifiable rewards, which limit their scalability and generality. Recent Self-Play RL methods, inspired by the success of the paradigm in g

Yusu Qian, Cheng Wan, Chao Jia, Yinfei Yang

Multimodal large language models (MLLMs) have achieved remarkable progress on vision-language tasks, yet their reasoning processes remain sometimes unreliable. We introduce PRISM-Bench, a benchmark of puzzle-based visual challenges designed to evaluate not only whether models can solve problems, but how their reasoning unfolds. Unlike prior evaluations that

Zhaoshu Yu, Bo Wang, Pengpeng Zeng, Haonan Zhang

Vision-Language-Action models (VLAs) represent a significant frontier in embodied intelligence, aiming to bridge digital knowledge with physical-world interaction. Despite their remarkable performance, foundational VLAs are hindered by the prohibitive computational and data demands inherent to their large-scale architectures. While a surge of recent research

Jiu-Yang He, Yu Jiang

The Terwilliger algebras of association schemes over an arbitrary field $\mathbb{F}$ were called the Terwilliger $\mathbb{F}$-algebras of association schemes in [9]. In this paper, we study the Terwilliger $\mathbb{F}$-algebras of factorial association schemes. We determine the centers, the semisimplicity, the Jacobson radicals and their nilpotent indices, t

Zhiping He, Xu Fang, Juejun Hu

Tunable active metasurfaces have become a major research focus in recent years. Among tuning mechanisms, all-optical coherent control stands out because it requires no material or geometric change, enabling ultrafast, low-energy, interference-based modulation of amplitude, phase, and polarization in ultrathin devices. However, when applied to phase-gradient

Matthew Kiser, Max McGinley, Daniel Malz

A limitation of analog quantum simulators based on cold atoms in optical lattices is that readout is typically limited to observables diagonal in the charge basis, i.e., densities and density correlation functions. To overcome this limitation, we propose experiment-friendly schemes to measure charge-off-diagonal correlations (such as currents). Our protocols

Cheol Woo Kim, Shresth Verma, Mauricio Tec, Milind Tambe

Direct Preference Optimization (DPO) has become a popular method for fine-tuning large language models (LLMs) due to its stability and simplicity. However, it is also known to be sensitive to noise in the data and prone to overfitting. Recent works have proposed using distributionally robust optimization (DRO) to address potential noise and distributional sh

Erich Liang, Roma Bhattacharjee, Sreemanti Dey, Rafael Moschopoulos

Accurately tracking camera intrinsics is crucial for achieving 3D understanding from 2D video. However, most 3D algorithms assume that camera intrinsics stay constant throughout a video, which is often not true for many real-world in-the-wild videos. A major obstacle in this field is a lack of dynamic camera intrinsics benchmarks--existing benchmarks typical

Guangting Zheng, Qinyu Zhao, Tao Yang, Fei Xiao

Directly modeling the explicit likelihood of the raw data distribution is key topic in the machine learning area, which achieves the scaling successes in Large Language Models by autoregressive modeling. However, continuous AR modeling over visual pixel data suffer from extremely long sequences and high-dimensional spaces. In this paper, we present FARMER, a

Yizhang Zhu, Liangwei Wang, Chenyu Yang, Xiaotian Lin

The rapid advancement of large language models (LLMs) has spurred the emergence of data agents, autonomous systems designed to orchestrate Data + AI ecosystems for tackling complex data-related tasks. However, the term "data agent" currently suffers from terminological ambiguity and inconsistent adoption, conflating simple query responders with sophisticated

Luis Ramos, Hiram Calvo, Olga Kolesnikova

The identification of hope speech has become a promised NLP task, considering the need to detect motivational expressions of agency and goal-directed behaviour on social media platforms. This proposal evaluates traditional machine learning models and fine-tuned transformers for a previously split hope speech dataset as train, development and test set. On dev

Feng-Li Lin, Sayid Mondal

We investigate whether Newtonian gravity can generate quantum entanglement between mesoscopic quantum bodies modeled as superposed mass quadrupoles using three complementary approaches: mini-superspace, semiclassical gravity, and stochastic gravity. We systematically analyze gravitationally induced entanglement (GIE) mechanisms and the conditions under which

Hong-Da Lyu, Jun-Kun Zhao, Li Li

Recent developments identify pole-skipping as a `smoking-gun' signature of the hydrodynamic nature of chaos, offering an alternative way to probe quantum chaos in addition to the out-of-time-ordered correlator (OTOC). We study the quantum chaos and pole-skipping phenomenon in the strongly coupled charged rotating fluids, holographically dual to rotating blac

R. Fernandez-Fernandez, M. M. Cueli, J. González-Nuevo, L. Bonavera

{Submillimetre galaxies (SMGs) are excellent background sources for magnification-bias studies, but the limited sky coverage in the submillimetre (sub-mm) band constrains their statistical power. Beyond H-ATLAS, Herschel produced additional sub-mm catalogues, though not optimised for spatial statistical lensing analyses.} {Our goal is to refine cosmological

Junyoung Seo, Rodrigo Mira, Alexandros Haliassos, Stella Bounareli

Audio-driven human animation models often suffer from identity drift during temporal autoregressive generation, where characters gradually lose their identity over time. One solution is to generate keyframes as intermediate temporal anchors that prevent degradation, but this requires an additional keyframe generation stage and can restrict natural motion dyn

Eric M. Schmid, Fernando Tohmé, William Chin

Let $Q$ be a finite quiver without oriented cycles, $C$ its path category, and $k$ a field. This paper is expository. We assemble in one place, and in elementary combinatorial form, the dictionary between Grothendieck topologies on $C$ and subcategories of $\mathrm{mod}\text{-}kQ$. Every Grothendieck topology on $C$ is rigid, so the sheaf categories are pres

Diego García, Sergio Arias, Rosa López

Quantum heat engines are nanoscale devices that convert heat into work by exploiting quantum effects, such as coherence and interference. Previous studies of these devices did not consider spin-dependent effects, which can influence the thermoelectric performance of the engine. In this work, we study the thermoelectric behavior of a quantum heat engine based

Joachim Baumann, Aleksandra Urman, Ulrich Leicht-Deobald, Zachary J. Roman

The rapid adoption of generative artificial intelligence (GenAI) technologies has led many organizations to integrate AI into their products and services, often without considering user preferences. Yet, public attitudes toward AI use, especially in impactful decision-making scenarios, are underexplored. Using a large-scale two-wave survey study (n_wave1=151

Anqi Li, Zhiyong Wang, Jiazhao Zhang, Minghan Li

Urban micromobility applications, such as delivery robots, demand reliable navigation across large-scale urban environments while following long-horizon route instructions. This task is particularly challenging due to the dynamic and unstructured nature of real-world city areas, yet most existing navigation methods remain tailored to short-scale and controll

Ulrik Enstad, Franz Luef

Bessel duality of regular Gabor systems states that a Gabor system over a lattice is a Bessel sequence if and only if the corresponding Gabor system over the adjoint lattice is a Bessel sequence. We show that this fundamental result of time-frequency analysis can be deduced from a theorem in the theory of bimodules over von Neumann algebras, namely that unde

Hongkai Lin, Dingkang Liang, Mingyang Du, Xin Zhou

Generative depth estimation methods leverage the rich visual priors stored in pre-trained text-to-image diffusion models, demonstrating astonishing zero-shot capability. However, parameter updates during training lead to catastrophic degradation in the image generation capability of the pre-trained model. We introduce MERGE, a unified model for image generat

Zihao Wang, Xujing Li, Yining Ye, Junjie Fang

We present Game-TARS, a generalist game agent trained with a unified, scalable action space anchored to human-aligned native keyboard-mouse inputs. Unlike API- or GUI-based approaches, this paradigm enables large-scale continual pre-training across heterogeneous domains, including OS, web, and simulation games. Game-TARS is pre-trained on over 500B tokens wi

P. M. Berto, F. CampodÓnico, A. A. Matoso, S. Vergara

Time-bin encoded photon pairs enable robust, decoherence-resistant transmission through optical fibers for long-distance quantum communication, where phase noise poses a critical limitation to stable operation. Here, we implement an adaptive Perturbation-and-Observe algorithm on a fully digital FPGA platform operating with real-time feedback at 1 Hz. The con

Yash Jangir, Yidi Zhang, Pang-Chi Lo, Kashu Yamazaki

The pursuit of robot generalists, agents capable of performing diverse tasks across diverse environments, demands rigorous and scalable evaluation. Yet real-world testing of robot policies remains fundamentally constrained: it is labor-intensive, slow, unsafe at scale, and difficult to reproduce. As policies expand in scope and complexity, these barriers onl

Thaís M. Dalbelo, Daniel Duarte, Danilo da Nóbrega Santos

In this work we present a formula for the Euler characteristic of the Milnor fiber of non-degenerate functions $f: X \to \mathbb{C}$ with isolated critical set relative to a stratification, where $X$ is a $2$-generic symmetric determinantal variety. The formula is obtained in two steps. Firstly, we explicitly describe the toric structure of those varieties.

Baoqi Pei, Yifei Huang, Jilan Xu, Yuping He

Egocentric video reasoning centers on an unobservable agent behind the camera who dynamically shapes the environment, requiring inference of hidden intentions and recognition of fine-grained interactions. This core challenge limits current multimodal large language models MLLMs, which excel at visible event reasoning but lack embodied, first-person understan

D. A. Dunne, K. A. Cleary, J. G. S. Lunde, D. T. Chung

We present a stack of data from the second season of the CO Mapping Array Project (COMAP) Pathfinder on the positions of quasars from eBOSS and DESI. COMAP is a Line Intensity Mapping (LIM) experiment targeting dense molecular gas via CO(1--0) emission at $z\sim3$. COMAP's Season 2 represents a $3\times$ increase in map-level sensitivity over the previous Ea

W. d'Assignies, G. M. Bernstein, B. Yin, G. Giannini

This work is part of a series establishing the redshift framework for the $3\times2$pt analysis of the Dark Energy Survey Year 6 (DES Y6). For DES Y6, photometric redshift distributions are estimated using self-organizing maps (SOMs), calibrated with spectroscopic and many-band photometric data. To overcome limitations from color-redshift degeneracies and in