May 2025 arXiv papers — page 28

Siddharth Ancha, Sunshine Jiang, Travis Manderson, Laura Brandt

In order to navigate safely and reliably in off-road and unstructured environments, robots must detect anomalies that are out-of-distribution (OOD) with respect to the training data. We present an analysis-by-synthesis approach for pixel-wise anomaly detection without making any assumptions about the nature of OOD data. Given an input image, we use a generat

Jonghan Lim, Ilya Kovalenko

Recent manufacturing systems are increasingly adopting multi-robot collaboration to handle complex and dynamic environments. While multi-agent architectures support decentralized coordination among robot agents, they often face challenges in enabling real-time adaptability for unexpected disruptions without predefined rules. Recent advances in large language

Pedro Mendes, Paolo Romano, David Garlan

Reliable uncertainty estimation is critical for deploying neural networks (NNs) in real-world applications. While existing calibration techniques often rely on post-hoc adjustments or coarse-grained binning methods, they remain limited in scalability, differentiability, and generalization across domains. In this work, we introduce CLUE (Calibration via Learn

Qing Wang, Yuepei Li, Qiao Qiao, Kang Zhou

Open Relation Extraction (OpenRE) seeks to identify and extract novel relational facts between named entities from unlabeled data without pre-defined relation schemas. Traditional OpenRE methods typically assume that the unlabeled data consists solely of novel relations or is pre-divided into known and novel instances. However, in real-world scenarios, novel

Abhishek Bajgain, Santu Prasad Jana, Alexander Samokhvalov, Thomas Parker

Scalable and controlled doping of two-dimensional transition metal dichalcogenides is essential for tuning their electronic and optoelectronic properties. In this work, we demonstrate a robust approach for substitution of vanadium in tungsten diselenide (WSe$_2$) via the selenization of pre-deposited V$_2$O$_5$/WO$_3$ thin films. By adjusting the thickness o

Anton Björklund, Mykola Zaitsev, Paolo Morettin, Marta Kwiatkowska

The growing reliance on artificial intelligence in safety- and security-critical applications is raising concerns about the robustness of neural networks to erroneous or adversarial input. Certification is a methodology for ensuring model trustworthiness by providing formal guarantees on model behaviour. While most verification methods focus on worst-case an

Vladyslav Gapyak, Thomas März, Andreas Weinmann

Magnetic Particle Imaging (MPI) is a promising tomographic technique for visualizing the spatio-temporal distribution of superparamagnetic nanoparticles, with applications ranging from cancer detection to real-time cardiovascular monitoring. Traditional MPI reconstruction relies on either time-consuming calibration (measured system matrix) or model-based sim

S. X. M. Riberolles, A. M. Nedić, B. Kuthanazhi, F. Ye

Topological properties of quantum materials are intimately related to symmetry. Here, we tune the magnetic order of the axion insulator candidate EuIn$_2$As$_2$ from its broken-helix ground state to the field-polarized phase by applying an in-plane magnetic field. Using results from neutron diffraction and magnetization measurements with ab inito theory and

Joe Karpie, Christopher J. Monahan, Kostas Orginos, Savvas Zafeiropoulos

Global analysis of collider and fixed-target experimental data and calculations from lattice quantum chromodynamics (QCD) are used to gain complementary information on the structure of hadrons. We propose novel ``window observables'' that allow for higher precision cross-validation between the different approaches, a critical step for studies that wish to co

Fidele J. Twagirayezu

We propose a novel timelike quantum energy teleportation (QET) protocol within the 1+1 dimensional Nambu-Jona-Lasinio (NJL) model, an interacting fermionic field theory exhibiting spontaneous chiral symmetry breaking. By coupling localized Unruh-DeWitt detectors to the fermionic field, we demonstrate how an initial observer's measurement enables a second obs

Srishti Yadav, Lauren Tilton, Maria Antoniak, Taylor Arnold

Modern vision-language models (VLMs) often fail at cultural competency evaluations and benchmarks. Given the diversity of applications built upon VLMs, there is renewed interest in understanding how they encode cultural nuances. While individual aspects of this problem have been studied, we still lack a comprehensive framework for systematically identifying

Yuxi Zhang, Yueting Li, Xinyu Du, Sibo Wang

Generating images from rhetorical languages remains a critical challenge for text-to-image models. Even state-of-the-art (SOTA) multimodal large language models (MLLM) fail to generate images based on the hidden meaning inherent in rhetorical language--despite such content being readily mappable to visual representations by humans. A key limitation is that c

Francesco Libbi, Lorenzo Monacelli, Boris Kozinsky

Similar to other perovskites in its family, $\mathrm{SrTiO_{3}}$ exhibits a significant softening of the ferroelectric mode with decreasing temperature, a behavior that typically heralds the onset of a ferroelectric transition. However, this material remains paraelectric down to 0K due to quantum fluctuations that prevent stabilization of the ferroelectric m

István Szanyi

The odderon, a leading crossing-odd $t$-channel exchange, was first proposed by L. Lukaszuk and B. Nicolescu in 1973, but its existence remained elusive for 48 years. Elastic proton-proton scattering measurements at CERN's Large Hadron Collider (LHC) and elastic proton-antiproton scattering measurements at FNAL's Tevatron, performed at TeV-scale center-of-ma

Erbing Hua, Theofilos Spyrou, Majid Ahmadi, Abdul Momin Syed

Memristor technology shows great promise for energy-efficient computing, yet it grapples with challenges like resistance drift and inherent variability. For filamentary Resistive RAM (ReRAM), one of the most investigated types of memristive devices, the expensive electroforming step required to create conductive pathways results in increased power and area o

Ghulam Mujtaba, Eun-Seok Ryu

EdgeVidSum is a lightweight method that generates personalized, fast-forward summaries of long-form videos directly on edge devices. The proposed approach enables real-time video summarization while safeguarding user privacy through local data processing using innovative thumbnail-based techniques and efficient neural architectures. Unlike conventional metho

Kristy Wedel

A critical challenge remains unresolved as generative AI systems are quickly implemented in various organizational settings. Despite significant advances in memory components such as RAG, vector stores, and LLM agents, these systems still have substantial memory limitations. Gen AI workflows rarely store or reflect on the full context in which decisions are

Christopher Polzak, Alejandro Lozano, Min Woo Sun, James Burgess

Systematic reviews (SR), in which experts summarize and analyze evidence across individual studies to provide insights on a specialized topic, are a cornerstone for evidence-based clinical decision-making, research, and policy. Given the exponential growth of scientific articles, there is growing interest in using large language models (LLMs) to automate SR

Xiang Liu, Junjie Wee, Guo-Wei Wei

Understanding how protein mutations affect protein-nucleic acid binding is critical for unraveling disease mechanisms and advancing therapies. Current experimental approaches are laborious, and computational methods remain limited in accuracy. To address this challenge, we propose a novel topological machine learning model (TopoML) combining persistent Lapla

Viraj Nadkarni, Pramod Viswanath

We present the first formal treatment of \emph{yield tokenization}, a mechanism that decomposes yield-bearing assets into principal and yield components to facilitate risk transfer and price discovery in decentralized finance (DeFi). We propose a model that characterizes yield token dynamics using stochastic differential equations. We derive a no-arbitrage p

Murthy L, Subarna Tripathi

The risk of misusing text-to-image generative models for malicious uses, especially due to the open-source development of such models, has become a serious concern. As a risk mitigation strategy, attributing generative models with neural fingerprinting is emerging as a popular technique. There has been a plethora of recent work that aim for addressing neural

Mytraya Gattu, J. K. Jain

One of the profound consequences of the fractional quantum Hall (FQH) effect is the notion of fractionally charged anyons. In spite of extensive experimental study, puzzles remain, however. For example, both shot-noise and Aharonov-Bohm interference measurements sometimes report a charge that is a multiple of the elementary charge. We report here high-precis

Antonio Ocello, Daniil Tiapkin, Lorenzo Mancini, Mathieu Laurière

We introduce Mean-Field Trust Region Policy Optimization (MF-TRPO), a novel algorithm designed to compute approximate Nash equilibria for ergodic Mean-Field Games (MFG) in finite state-action spaces. Building on the well-established performance of TRPO in the reinforcement learning (RL) setting, we extend its methodology to the MFG framework, leveraging its

Hammed Oyekan

Accurate depth estimation of magnetic sources plays a crucial role in various geophysical applications, including mineral exploration, resource assessments, regional hydrocarbon exploration, and geological mapping. Thus, this abstract presents a fast and simple method of estimating the depth of a magnetic body using the TDX derivative of the total magnetic f

Mohammad Helal Uddin, Sabur Baidya

Mental disorders including depression, anxiety, and other neurological disorders pose a significant global challenge, particularly among individuals exhibiting social avoidance tendencies. This study proposes a hybrid approach by leveraging smartphone sensor data measuring daily physical activities and analyzing their social media (Twitter) interactions for

Sarah Meiklejohn, Hayden Blauzvern, Mihai Maruseac, Spencer Schrock

Powerful machine learning (ML) models are now readily available online, which creates exciting possibilities for users who lack the deep technical expertise or substantial computing resources needed to develop them. On the other hand, this type of open ecosystem comes with many risks. In this paper, we argue that the current ecosystem for open ML models cont

John Mendonça, Alon Lavie, Isabel Trancoso

Evaluating the quality of open-domain chatbots has become increasingly reliant on LLMs acting as automatic judges. However, existing meta-evaluation benchmarks are static, outdated, and lacking in multilingual coverage, limiting their ability to fully capture subtle weaknesses in evaluation. We introduce MEDAL, an automated multi-agent framework for curating

Merlijne Geurts, Tren Baltussen, Alexander Katriniok, Maurice Heemels

This research introduces a multi-horizon contingency model predictive control (CMPC) framework in which classes of robust MPC (RMPC) algorithms are combined with classes of learning-based MPC (LB-MPC) algorithms to enable safe learning. We prove that the CMPC framework inherits the robust recursive feasibility properties of the underlying RMPC scheme, thereb

Arka Banerjee, Abhik Bhattacharjee, Subinoy Das, Anshuman Maharana

In this work we present Fisher forecasts on \textit{non-thermal LiMR} models for a CMB Stage IV-like experiment and the Simons Observatory -- particularly focusing on a model of inflaton/moduli decay giving rise to non-thermally distributed dark sector particles, and also comparing our results with those for sterile particles following the Dodelson-Widrow di

Kaja Dobrovoljc

This paper presents a novel treebank-driven approach to comparing syntactic structures in speech and writing using dependency-parsed corpora. Adopting a fully inductive, bottom-up method, we define syntactic structures as delexicalized dependency (sub)trees and extract them from spoken and written Universal Dependencies (UD) treebanks in two syntactically di

Rostyslav O. Serha, Kaitlin H. McAllister, Fabian Majcen, Sebastian Knauer

Solid-state platforms based on bosonic quasiparticles offer a compelling route toward on-chip quantum information technologies scalable to nanometer dimensions. Coherence time, a key figure of merit for any quantum system, is fundamentally limited by the lifetime of quasiparticles that store quantum information. For magnons - bosonic excitations of collectiv

Claas Voelcker, Anastasiia Pedan, Arash Ahmadian, Romina Abachi

The idea of value-aware model learning, that models should produce accurate value estimates, has gained prominence in model-based reinforcement learning. The MuZero loss, which penalizes a model's value function prediction compared to the ground-truth value function, has been utilized in several prominent empirical works in the literature. However, theoretic

Jiahui Zheng, Cole Jahnke, Wei "Wayne" Chen

This paper introduces GUST (Generative Uncertainty learning via Self-supervised pretraining and Transfer learning), a framework for quantifying free-form geometric uncertainties inherent in the manufacturing of metamaterials. GUST leverages the representational power of deep generative models to learn a high-dimensional conditional distribution of as-fabrica

Canh Thien Dang, An Nguyen

As artificial intelligence (AI) becomes integral to the society, the ability to critically evaluate AI-generated content is increasingly vital. On the context of management education, we examine how academic skills, cognitive traits, and AI scepticism influence students' ability to detect factually incorrect AI-generated responses (hallucinations) in a high-

Christopher Ormerod

This study illustrates how incorporating feedback-oriented annotations into the scoring pipeline can enhance the accuracy of automated essay scoring (AES). This approach is demonstrated with the Persuasive Essays for Rating, Selecting, and Understanding Argumentative and Discourse Elements (PERSUADE) corpus. We integrate two types of feedback-driven annotati

Arjun Rao, Hanieh Alipour, Nick Pendar

This paper presents a comparison of embedding models in tri-modal hybrid retrieval for Retrieval-Augmented Generation (RAG) systems. We investigate the fusion of dense semantic, sparse lexical, and graph-based embeddings, focusing on the performance of the MiniLM-v6 and BGE-Large architectures. Contrary to conventional assumptions, our results show that the

Iacopo Nonis

Let $k$ be an algebraically closed field. Let $R$ be a local commutative finite dimensional $k$-algebra and let $Q$ be a quiver with no loops or oriented cycles. We show that mutation of $\tau$-exceptional sequences over $\Lambda = R\otimes_k kQ \cong RQ$ in the sense of Buan, Hanson, and Marsh coincides with the classical mutation of exceptional sequences d

Yaxiong Lei, Mingyue Zhao, Yuheng Wang, Shijing He

Mobile gaze tracking faces a fundamental challenge: maintaining accuracy as users naturally change their postures and device orientations. Traditional calibration approaches, like one-off, fail to adapt to these dynamic conditions, leading to degraded performance over time. We present MAC-Gaze, a Motion-Aware continual Calibration approach that leverages sma

Mert Onur Cakiroglu, Idil Bilge Altun, Mehmet Dalkilic, Elham Buxton

Time series forecasting remains a challenging task for foundation models due to temporal heterogeneity, high dimensionality, and the lack of inherent symbolic structure. In this work, we propose DRAGON (Discrete Representation and Augmented Graph encoding Over de BruijN Graphs), a novel encoder that introduces Multivariate de Bruijn Graphs (MdBGs) to bridge

Eleni Vasilaki

Large Language Models (LLMs) can be understood as Collective Knowledge (CK): a condensation of human cultural and technical output, whose apparent intelligence emerges in dialogue. This perspective article, drawing on extended interaction with ChatGPT-4, postulates differential response modes that plausibly trace their origin to distinct model subnetworks. I

Paula M. S. Fialho, Aldo Procacci

We present two new contributions to the study of the independence polynomial $Z_G(z)$ of a finite simple graph $G = (V,E)$. First, we provide an improved lower bound for the zero-free region of $Z_G(z)$ for the important class of claw-free graphs. Our bound exceeds the classical Shearer radius and it is derived through a refined application of the Fern\'ande

Iddo Yosha, Gallil Maimon, Yossi Adi

Sentence stress refers to emphasis on words within a spoken utterance to highlight or contrast an idea. It is often used to imply an underlying intention not explicitly stated. Recent speech-aware language models (SLMs) have enabled direct audio processing, allowing models to access the full richness of speech to perform audio reasoning tasks such as spoken

Divya Shanmugam, Helen Lu, Swami Sankaranarayanan, John Guttag

A conformal classifier produces a set of predicted classes and provides a probabilistic guarantee that the set includes the true class. Unfortunately, it is often the case that conformal classifiers produce uninformatively large sets. In this work, we show that test-time augmentation (TTA)--a technique that introduces inductive biases during inference--reduc

Mark P. Hertzberg, Daniel Jiménez-Aguilar

Some recent studies based on numerical relativity simulations claim that slow contraction/ekpyrosis is strongly preferred over inflation as the smoothing mechanism that brought the universe into the homogeneous, isotropic and flat state we observe today on large scales. In this paper, we evaluate the likelihood of the initial conditions employed in the afore

Marco Colussi, Dragan Ahmetovic, Sergio Mascetti

This paper presents MIAS-SAM, a novel approach for the segmentation of anomalous regions in medical images. MIAS-SAM uses a patch-based memory bank to store relevant image features, which are extracted from normal data using the SAM encoder. At inference time, the embedding patches extracted from the SAM encoder are compared with those in the memory bank to

Afila Ajithkumar Sophiya, Akarsh K Nair, Sepehr Maleki, Senthil K. Krishnababu

Physics Informed Neural Networks (PINNs) have been emerging as a powerful computational tool for solving differential equations. However, the applicability of these models is still in its initial stages and requires more standardization to gain wider popularity. Through this survey, we present a comprehensive overview of PINNs approaches exploring various as

Razvan-Andrei Lascu, Mateusz B. Majka

We study the problem of minimizing non-convex functionals on the space of probability measures, regularized by the relative entropy (KL divergence) with respect to a fixed reference measure, as well as the corresponding problem of solving entropy-regularized non-convex-non-concave min-max problems. We utilize the Best Response flow (also known in the literat

Sara Papi, Marco Gaido, Luisa Bentivogli, Alessio Brutti

The development of speech foundation models (SFMs) like Whisper and SeamlessM4T has significantly advanced the field of speech processing. However, their closed nature--with inaccessible training data and code--poses major reproducibility and fair evaluation challenges. While other domains have made substantial progress toward open science by developing full

Aniruddha Nrusimha, William Brandon, Mayank Mishra, Yikang Shen

The size and compute characteristics of modern large language models have led to an increased interest in developing specialized kernels tailored for particular training and inference workloads. Existing kernels primarily optimize for compute utilization, targeting the large-batch training and inference settings. However, low-batch inference, where memory ba

Ansar Aynetdinov, Alan Akbik

Multi-token prediction (MTP) is a recently proposed pre-training objective for language models. Rather than predicting only the next token (NTP), MTP predicts the next $k$ tokens at each prediction step, using multiple prediction heads. MTP has shown promise in improving downstream performance, inference speed, and training efficiency, particularly for large

Tian Qin, Core Francisco Park, Mujin Kwun, Aaron Walsman

Mathematical reasoning tasks have become prominent benchmarks for assessing the reasoning capabilities of LLMs, especially with reinforcement learning (RL) methods such as GRPO showing significant performance gains. However, accuracy metrics alone do not support fine-grained assessment of capabilities and fail to reveal which problem-solving skills have been

Z. Sela

The JSJ decomposition and the Makanin-Razborov diagram were proved to be essential in studying varieties over free groups, semigroups and associative algebras. In this paper we suggest a unified conceptual approach to the applicability of these structures over all these algebraic categories. With a variety over each of these algebraic categories we naturally

P. S. Bhupal Dev, Bhaskar Dutta, Aparajitha Karthikeyan, Writasree Maitra

The recent KM3NeT observation of an ${\cal{O}}(100~{\rm PeV})$ event KM3-230213A is puzzling because IceCube with much larger effective area times exposure has not found any such events. We propose a novel solution to this conundrum in terms of dark matter (DM) scattering in the Earth's crust. We show that intermediate dark-sector particles that decay into m

Arseniy Pertzovsky, Roni Stern, Ariel Felner, Roie Zivan

We explore the use of Artificial Potential Fields (APFs) to solve Multi-Agent Path Finding (MAPF) and Lifelong MAPF (LMAPF) problems. In MAPF, a team of agents must move to their goal locations without collisions, whereas in LMAPF, new goals are generated upon arrival. We propose methods for incorporating APFs in a range of MAPF algorithms, including Priorit

Rafik Mankour, Yassine Chafai, Hamada Saleh, Ghassen Ben Hassine

Climate Finance Bench introduces an open benchmark that targets question-answering over corporate climate disclosures using Large Language Models. We curate 33 recent sustainability reports in English drawn from companies across all 11 GICS sectors and annotate 330 expert-validated question-answer pairs that span pure extraction, numerical reasoning, and log

Bingqing Hu, Bin Nan

Traditional survival models often rely on restrictive assumptions such as proportional hazards or instantaneous effects of time-varying covariates on the hazard function, which limit their applicability in real-world settings. We consider the nonparametric estimation of the conditional survival function, which leverages the flexibility of neural networks to

Shweta Sharma, Abhishek Sharma, Abhinendra Singh

Dense particulate suspensions often exhibit a dramatic increase in viscosity in response to external deformation. This shear thickening behavior has been related to a transition from lubricated, unconstrained pairwise motion to a frictional contact network (FCN) at high stresses. Here, we study the characteristics of the FCN formed during shear thickening to

Jose Guadalupe Hernandez, Attri Ghosh, Philip J. Freda, Yufei Meng

We present the Star-Based Automated Single-locus and Epistasis analysis tool - Genetic Programming (StarBASE-GP), an automated framework for discovering meaningful genetic variants associated with phenotypic variation in large-scale genomic datasets. StarBASE-GP uses a genetic programming-based multi-objective optimization strategy to evolve machine learning

Mohit Raj Sah, Akash Maurya, Suvodip Mukherjee, Prayush Kumar

The stochastic gravitational wave background (SGWB) in the nanohertz (nHz) regime, detectable by pulsar timing arrays (PTAs), provides a promising probe of the cosmic population of supermassive black hole binaries (SMBHBs). These binaries are expected to retain significant eccentricity throughout their evolution. We present a new technique to model the nHz S

Aycke Roos, Pablo M. Maier, Andres F. Ordonez, Olga Smirnova

We develop a mathematical formalism underlying the emergence of enantio-sensitive molecular orientation due to photoionization or photoexitation of chiral molecules. We consider geometric quantities such as the Berry connection and Berry curvature in light-driven chiral electronic states in the space of complex light polarization vectors. The parametric depe

Sitan Chen, Weiyuan Gong, Jonas Haferkamp, Yihui Quek

In a variety of physically relevant settings for learning from quantum data, designing protocols that can computationally efficiently extract information remains largely an art, and there are important cases where we believe this to be impossible, that is, where there is an information-computation gap. While there is a large array of tools in the classical l

Debosmita Pathak, Charlotte R. Christensen, Alyson M. Brooks, Ferah Munshi

It is necessary to understand the full accretion history of the Milky Way in order to contextualize the properties of observed Milky Way satellite galaxies and the stellar halo. This paper compares the dynamical properties and star-formation histories of surviving and disrupted satellites around Milky Way-like galaxies using the DC Justice League suite of ve

Evan Peters

Data-driven decoding (DDD) - learning to decode syndromes of (quantum) error-correcting codes by learning from data - can be a difficult problem due to several atypical and poorly understood properties of the training data. We introduce a theory of example importance that clarifies these unusual aspects of DDD: For instance, we show that DDD of a simple erro

Marvin Gerlach, Ulrich Nierste, Pascal Reeck, Vladyslav Shtabovenko

We compute next-to-next-to-leading order perturbative corrections to the decay width difference of mass eigenstates and the charge-parity asymmetry $a_{\rm fs}$ in flavour-specific decays of neutral $B$ mesons. In our calculation we take into account the full dependence on the charm and bottom quark masses for the current-current operator contributions up to

Samsuzzaman Afroz, Suvodip Mukherjee

Gravitational wave (GW) observations have significantly advanced our understanding of binary compact object (BCO) formation, yet directly linking these observations to specific formation scenarios remains challenging. The BCO phase space provides a robust and data-driven approach to discover the likely formation scenarios of these binaries. In this study, we

Paul L. Schechter, Jeremy D. Schnittman

Even when used to describe the same phenomenon, equations, graphics and words each give different perspectives and lead to complementary insights. The basic elements of strong gravitational lensing are introduced here favoring words and graphics over equations whenever possible. Fermat's principle is the fundamental driver of strong lensing. Three "D's'' enc

Takumi Matsuzawa, Minhui Zhu, Nigel Goldenfeld, William T. M. Irvine

Turbulence, left unforced, decays and invades the surrounding quiescent fluid. Though ubiquitous, this simple phenomenon has proven hard to capture within a simple and general framework. Experiments in conventional turbulent flow chambers are inevitably complicated by proximity to boundaries and mean flow, obscuring the fundamental aspects of the relaxation

Alireza Allahyari, Mohammadreza Davari, David F. Mota

We explore a theoretical framework in which Lorentz symmetry is explicitly broken by incorporating derivative terms of the extrinsic curvature into the gravitational action. These modifications introduce a scale-dependent damping effect in the propagation of gravitational waves (GWs), governed by a characteristic energy scale denoted as $M_{{LV}}$ . We deriv

Tong Sun, Bowen Jiang, Hailong Lin, Borui Li

To safeguard user data privacy, on-device inference has emerged as a prominent paradigm on mobile and Internet of Things (IoT) devices. This paradigm involves deploying a model provided by a third party on local devices to perform inference tasks. However, it exposes the private model to two primary security threats: model stealing (MS) and membership infere

Brandon Barton, Juan Carrasquilla, Christopher Roth, Agnes Valenti

The Lottery Ticket Hypothesis (LTH) posits that within overparametrized neural networks, there exist sparse subnetworks that are capable of matching the performance of the original model when trained in isolation from the original initialization. We extend this hypothesis to the unsupervised task of approximating the ground state of quantum many-body Hamilto

Lachlan Lancaster, Chang-Goo Kim, Jeong-Gyu Kim, Eve C. Ostriker

In a companion paper (Paper I) we presented a Co-Evolution Model (CEM) in which to consider the evolution of feedback bubbles driven by massive stars through both stellar winds and ionizing radiation, outlining when either of these effects is dominant and providing a model for how they evolve together. Here we present results from three-dimensional radiation

Joonas Hirvonen, Oliver Gould

We show that Langer's rate of bubble nucleation is quantitatively correct up to small higher-loop corrections, in comparison to lattice simulations. These results are a significant advancement on decades of lattice studies showing only qualitative trends, and the first showing agreement for any conservative system. We confirm that the failure to fully therma

Andrei Tsypilnikov, Matheus Fibger, Fernando Iemini

We discuss the behavior of general Floquet time crystals (FTCs), including prethermal ones, in closed systems acting as AC sensors. We provide an analytical treatment of their quantum Fisher information (QFI) dynamics, which characterizes the ultimate sensor accuracy. By tuning the direction and frequency of the AC field, we show how to induce transitions re

Lachlan Lancaster, Jeong-Gyu Kim, Greg L. Bryan, Shyam H. Menon

We propose a new framework for the simultaneous feedback of stellar winds and photo-ionizing radiation from massive stars, distinguishing the locations where forces are applied, and consequences for internal spatio-temporal evolution of the whole feedback bubble (FB). We quantify the relative dynamical importance of wind-blown bubbles (WBB) versus the photoi

Ted M. Johnson, Rebecca G. Martin, Stephen Lepp, Stephen H. Lubow

Circumbinary gas disks that are misaligned to the binary orbital plane evolve toward either a coplanar or a polar-aligned configuration with respect to the binary host. The preferred alignment depends on the dynamics of the disk: whether it undergoes librating or circulating nodal precession, with librating disks evolving to polar inclinations and circulatin

Pavel E. Mancera Piña, Justin I. Read, Stacy Kim, Antonino Marasco

(Abridged) The relations between stellar ($M_\ast$), gas ($M_{\rm gas}$), baryonic ($M_{\rm bar} = M_\ast + M_{\rm gas}$), and dark matter halo mass ($M_{200}$) provide unique constraints on galaxy formation and cosmology. The shape of the relations constrains how galaxies regulate their growth through gas accretion, star formation, and feedback; their scatt

Kiril Hristov, Peng-Ju Hu, Yi Pang

We study Einstein-Maxwell theory in $D \geq 3$ spacetime dimensions including all Lorentz-invariant parity-even four-derivative couplings. Building on the results of arXiv:2312.11610, we consider static, charged, asymptotically flat black hole solutions to first order in the higher-derivative expansion. In $D=4$ and $D=5$, we compute the corrected black hole

Jonas Beck, Jonathan Bodky, Matteo Dürrnagel, Ronny Thomale

Translation symmetry-breaking order is assumed to be suppressed by the lack of Fermi surface nesting near certain higher-order Van Hove singularities (HOVHS). We show the anisotropic band-flattening inherent to such HOVHS, combined with broadening of the Fermi surface due to elevated critical temperatures, results in the Fermi surface becoming approximately

Ralph E. Pudritz, Alex J. Cridland, Julie Inglis, Mathew Alessi

Extensive ground and space based surveys have now characterized the properties of thousands of exoplanets; their radii, masses, orbits around their host stars, and the beginnings of accurate measurements of the chemical compositions of their atmospheres and cores. How are these properties linked to their formation in physically and chemically evolving protop

Max Uetrecht, Igor F. Herbut, Michael M. Scherer, Emmanuel Stamou

Two-dimensional materials with interacting Dirac excitations can host quantum multicritical behavior near the phase boundaries of the semimetallic and two-ordered phases. We study such behavior in Gross--Neveu--Yukawa field theories where $N_f$ flavors of Dirac fermions are coupled to two order-parameter fields with $SO(N_A)$ and $SO(N_B)$ symmetry, respecti

Michael J. Baker, Joaquim Iguaz Juan, Aidan Symons, Andrea Thamm

The KM3NeT experiment has recently observed a neutrino with an energy around 100PeV, and IceCube has detected five neutrinos with energies above 1PeV. While there are no known astrophysical sources, exploding primordial black holes could have produced these high-energy neutrinos. For Schwarzschild black holes this interpretation results in tensions between t

Naushad A. Kamar, Mostafa Ali, Mohammad Maghrebi

Quantum phase transitions are a cornerstone of many-body physics at low temperatures but have remained elusive far from equilibrium. Driven open quantum systems -- a prominent non-equilibrium platform where coherent dynamics competes with Markovian dissipation from the environment -- often exhibit an effective classical behavior. In this work, we present a n

Malte Pütz, Romain Vasseur, Andreas W. W. Ludwig, Simon Trebst

In circuit-based quantum state preparation, qubit loss and coherent errors are circuit imperfections that imperil the formation of long-range entanglement beyond a certain threshold. The critical theory at the threshold is a continuous entanglement transition known to be described by a (2+0)-dimensional non-unitary conformal field theory which, for the two t

Jeff A. Dror, Qiushi Wei

Ultralight vector dark matter induces metric fluctuations that generate timing residuals in the arrival times of pulsar emissions through two distinct modes: a fast mode, sourced by coherent field oscillations, and a slow mode, arising from interference patterns. These modes enable the detection of vector dark matter with masses $m \sim 10^{-24} - 10^{-22}\

Laura Reymond, Alexander Reeves, Pierre Zhang, Alexandre Refregier

The upcoming stage IV wide-field surveys will provide high precision measurements of the large-scale structure (LSS) of the universe. Their interpretation requires fast and accurate theoretical predictions including large scales. For this purpose, we introduce $\texttt{SwiftC}_\ell$, a fast, accurate and differentiable $\texttt{JAX}$-based pipeline for the c

Raphaël Errani, Jorge Peñarrubia, Matthew G. Walker

We study the effect of stellar mass segregation driven by collisional relaxation within the potential well of a smooth dark matter halo. This effect is of particular relevance for old stellar systems with short crossing times, where small collisional perturbations accumulate over many dynamical timescales. We run collisional $N$-body simulations tailored to

Sergio E. Aguilar-Gutierrez

In this work, we formulate the holographic dictionary for the double-scaled SYK (DSSYK) model with matter operators. Based on the two-sided Hartle-Hawking (HH) state, we derive several properties of the DSSYK model, without making assumptions about the specific dual theory, including its semiclassical thermodynamics, correlation functions, and Krylov complex

Yannick Stade, Wan-Hsuan Lin, Jason Cong, Robert Wille

Quantum computing promises to solve previously intractable problems, with neutral atoms emerging as a promising technology. Zoned neutral atom architectures allow for immense parallelism and higher coherence times by shielding idling atoms from interference with laser beams. However, in addition to hardware, successful quantum computation requires sophistica

Rachit Kapoor, J. Ramirez Diaz, D. Yahne, V. O. Garlea

Thermodynamic and muon spin-relaxation measurements have recently highlighted BiYbGeO$_5$ as a new example of a rare-earth-based quantum dimer magnet with isolated Yb$^{3+}$ spin-$\frac{1}{2}$ dimers. However, direct spectroscopic evidence of the triplet excitations and measurements of the structural disorder are lacking. In this work, polycrystalline BiYbGe

Matthew Reece, Tom Rudelius, Christopher Tudball

Towers of electrically and magnetically charged states in quantum gravity often exhibit two important properties. First, the ratio of the mass (or tension) of electrically charged states to magnetically charged states is of order $e^2/(4\pi)$, which we refer to as "co-scaling." Second, in theories of multiple gauge fields, the towers of states that exhibit c

Andrew Marszewski, Claude-André Faucher-Giguère, Robert Feldmann, Guochao Sun

Recent observations suggest a nearly constant gas-phase mass-metallicity relation (MZR) at $z \gtrsim 5$, in agreement with many theoretical predictions. This lack of evolution contrasts with observations at $z \lesssim 3$, which find an increasing normalization of the MZR with decreasing redshift. We analyze a high-redshift suite of FIRE-2 cosmological zoom

Yasaman Farzan, Matheus Hostert

We entertain the possibility that transient astrophysical sources can produce a flux of dark particles that induce ultra-high-energy signatures at neutrino telescopes such as IceCube and KM3NeT. We construct scenarios where such ``dark flux" can produce meta-stable dark particles inside the Earth that subsequently decay to muons, inducing through-going track

Pouya Asadi, Marianne Moore, David E. Morrissey, Michael Shamma

We introduce a new mechanism for the simultaneous generation of baryon and dark matter asymmetries through ultraviolet-dominated freeze-in scatterings. The mechanism relies on heavy Majorana neutrinos that connect the visible Standard Model sector to a dark sector through the neutrino portal. Following reheating of the visible sector to a temperature well be

Kirk S. S. Barrow, Thinh Huu Nguyen, Edward C. Skrabacz

We describe a new Python-based stand-alone halo finding algorithm, Haskap Pie, that combines several methods of halo finding and tracking into a single calculation. Our halo-finder flexibly solves halos for simulations produced by eight simulation codes (ART-I, ENZO, RAMSES, CHANGA, GADGET-3, GEAR, AREPO, and GIZMO) and for both zoom-in or full-box N-body or

Steven B. Giddings

The questions of describing observables and observation in quantum gravity appear to be centrally important to its physics. A relational approach holds significant promise, and a classification of different types of relational observables (gravitationally dressed, field relational, and more general) is outlined. Plausibly gravitationally dressed observables

Fabio Rigamonti, Lorenzo Bertassi, Riccardo Buscicchio, Fabiola Cocchiararo

Supermassive black hole binaries (SBHBs) are a natural outcome of galaxy mergers, and they are expected to be among the loudest gravitational-wave sources at low frequencies. The source SDSS J2320+0024 was recently proposed as a promising SBHB candidate due to a possible periodicity in its light curve and variability in the MgII emission line. In this work,

Kaiyu Yue, Vasu Singla, Menglin Jia, John Kirchenbauer

Vision language models (VLMs) typically pair a modestly sized vision encoder with a large language model (LLM), e.g., Llama-70B, making the decoder the primary computational burden during training. To reduce costs, a potential promising strategy is to first train the vision encoder using a small language model before transferring it to the large one. We cons

Aimon Rahman, Kartik Narayan, Vishal M. Patel

Stylized abstraction synthesizes visually exaggerated yet semantically faithful representations of subjects, balancing recognizability with perceptual distortion. Unlike image-to-image translation, which prioritizes structural fidelity, stylized abstraction demands selective retention of identity cues while embracing stylistic divergence, especially challeng

Feng Luo, Yu-Neng Chuang, Guanchu Wang, Hoang Anh Duy Le

Reasoning-capable large language models (LLMs) achieve strong performance on complex tasks but often exhibit overthinking after distillation, generating unnecessarily long chain-of-thought (CoT) reasoning even for simple inputs and incurring high inference cost. However, naively shortening reasoning length can degrade reasoning accuracy, as concise reasoning

Qingchen Yu, Zifan Zheng, Ding Chen, Simin Niu

The evaluation of large language models (LLMs) has traditionally relied on static benchmarks, a paradigm that poses two major limitations: (1) predefined test sets lack adaptability to diverse application domains, and (2) standardized evaluation protocols often fail to capture fine-grained assessments of domain-specific knowledge and contextual reasoning abi