April 2026 arXiv papers — page 125

Alexander Osinsky, Ivan Kozyrev

The paper considers the problem of finding a submatrix $X_{\mathcal{S}} \in \mathbb{R}^{m \times k}$ in a matrix $X \in \mathbb{R}^{m \times n}$, such that the spectral or Frobenius norm of $X_{\mathcal{S}}^{\dag} X$ is limited, which guarantees it provides a good representation of the whole matrix. Such bounds can be reached by applying greedy algorithms, m

Jen Rogers, Derya Akbaba, James Scott-Brown, Alexander Lex

Decades of advocacy for reproducibility and replication have advanced open, transparent practices in the sciences. However, traditional notions of reproducibility fit poorly with design-oriented visualization research, where insights emerge through subjective, situated, and iterative work. So how can we ensure rigor and transparency in processes that are inh

Todd Hildebrant

We study independent sets in strong powers of circulant graphs using a transfer matrix formulation. The compatibility constraints separate into intra-layer and inter-layer components, yielding a transfer operator that is equivariant under the dihedral group action. The characteristic polynomial of the transfer operator factors into an \emph{anomalous} compon

Joel Kuttruff, Ritwika Mandal, Marina Servol, Céline Mariette

Correlated transition metal oxides present exciting prospects as switches or memory and storage devices owing to the possibility to control electronic properties using various external stimuli. While their complex behaviour is known to stem from interplay between electronic correlations, atomic structure and orbital physics, they remain poorly understood on

Ferdinand M. Schessl

Turn-level metrics are widely used to evaluate properties of multi-turn human-LLM conversations, from safety and sycophancy to dialogue quality. However, consecutive turns within a conversation are not statistically independent -- a fact that virtually all current evaluation pipelines fail to correct for in their statistical inference. We systematically char

Alexei Rybkin

We consider the Cauchy problem for the Korteweg--de Vries equation with real initial data $q$ that is both $L^1$ and $L^2$ summable and supported on (0,\infty). Using the left reflection coefficient and Hankel operators on the Hardy space $H^2$, we derive a trace-type representation for the corresponding solution. The proof is based on approximation by compa

Marco Ronzani, Cristina Silvano

Hypergraph partitioning is a pervasive NP-hard problem, and accelerating its computation on GPU can both slice time-to-solution and raise quality of results. In this work, we implement a multi-level hypergraph partitioning algorithm on GPU targeting a specific set of problem constraints: bounded per-partition size and distinct inbound hyperedges. Manipulatin

Mauricio Bustamante, Qinrui Liu, Gabriela Barenboim

While conventional oscillation experiments measure neutrino mixing parameters with high precision, these measurements are strictly confined to sub-TeV scales. At higher energies, renormalization-group effects can cause these parameters to evolve with the transferred momentum, $Q$. High-energy and ultra-high-energy astrophysical neutrinos, spanning TeV to EeV

Keon Kim, Krish Chelikavada

Multi-step zoom-in pipelines are widely used for GUI grounding, yet the intermediate predictions they produce are typically discarded after coordinate remapping. We observe that these intermediate outputs contain a useful confidence signal for free: zoom consistency, the distance between a model's step-2 prediction and the crop center. Unlike log-probabiliti

MD Awsaf Alam Anindya, Showvik Biswas, Anindya Iqbal, Jaydeb Sarker

Toxic interactions during code reviews can undermine teamwork and hinder productivity in software engineering (SE) teams. While prior studies explore toxicity detection and empirical investigation, they lack real-time detoxification tools to support the SE community. To address this gap, we present ToxiShield, a browser extension for GitHub pull requests tha

Emma K. Mackay, Amol A. Verma, Fahad Razak, Surain B. Roberts

Propensity score weighting approaches have been widely implemented in clinical research to estimate the effects of a treatment or exposure while mitigating the risk of confounding in the absence of random assignment. In practice, when working with large electronic health records (EHR) or administrative datasets to evaluate health quality outcomes at the inst

Joseph Walton, Gabriel Nicolosi

While reinforcement learning has been increasingly applied to stochastic control, few studies have systematically examined policy-based methods in queuing environments modeled as a semi-Markov decision process (SMDP). To address this gap, we investigate how policy-based reinforcement learning (RL) algorithms perform when applied to the control of service rat

Csaba Biró, Caroline E. Boone, Beth Novick, Hazel Torek

We study the metric dimension (strong and weak) of infinite graphs. In particular, our main interest is characterizing infinite graphs with finite dimension. Our main results: (1) graphs with more than one end have infinite strong dimension; (2) for graphs with a finite number of cycles, the weak dimension is finite if and only if the graph has finitely many

Ilsang Ohn, Shitao Fan, Jungbin Jun, Lizhen Lin

When considering a model selection or, more generally, an aggregation approach for adaptive statistical inference, it is often necessary to compute estimators over a wide range of model complexities including unnecessarily large models even when the true data-generating process is relatively simple, due to the lack of prior knowledge. This requirement can le

Julian Killingback, Ofer Meshi, Henry Li, Hamed Zamani

Traditional Retrieval-Augmented Generation (RAG) approaches generally assume that retrieval and generation occur on powerful servers removed from the end user. While this reduces local hardware constraints, it introduces significant drawbacks: privacy concerns regarding data access, recurring maintenance and storage costs, increased latency, and the necessit

Mikhail V. Medvedev, Anatoly Spitkovsky, Alexander Philippov

The guiding center formalism is employed to analyze the motion of a charged relativistic particle in an inhomogeneous magnetic field, subject to magnetic mirroring and energy loss due to cooling. The governing equation for the evolution of the magnetic moment is derived. An example representing a neutron star (pulsar or magnetar) magnetosphere is presented t

Duo Lu, Andrew Crotty, Uğur Çetintemel

Agentic AI systems are becoming commonplace in domains that require long-lived, stateful decision-making in continuously evolving conditions. As such, correctness depends not only on the output of individual model calls, but also on how to best adapt when incorporating new evidence or revising prior conclusions. However, existing frameworks rely on imperativ

Bingwei Zhang, Thomas Chen, Kai Hormann, Chee Yap

Range functions are a fundamental tool for certified computations in geometric modeling, computer graphics, and robotics, but traditional range functions have only quadratic convergence order ($m=2$). For ``superior'' convergence order (i.e., $m>2$), we exploit the Cornelius--Lohner framework in order to introduce new bivariate range functions based on Taylo

Aodi Wu, Haodong Han, Xubo Luo, Ruisuo Wang

Autonomous on-orbit servicing demands embodied agents that perceive through visual sensors, reason about 3D spatial situations, and execute multi-phase tasks over extended horizons. We present SpaceMind, a modular and self-evolving vision-language model (VLM) agent framework that decomposes knowledge, tools, and reasoning into three independently extensible

Kristian Holme, Jean Rabault, Ricardo Vinuesa, Mikael Mortensen

Rotating detonation engines (RDEs) are a promising propulsion concept that may offer higher thermodynamic efficiency and specific impulse than conventional systems, but nonlinear phenomena, including transitions to oscillatory or chaotic propagation modes, can hinder practical operation. Deep Reinforcement Learning (DRL) has emerged as a promising method for

Alexander Iksanov, Oleh Iksanov

Using a probabilistic technique based on exponential change of measure, we derive precise tail asymptotics of some perpetuities with distributions close to the Dickman distribution.

Rajendra Dulal, Serafim Teknowijoyo, Sara Chahid, Vahan Nikoghosyan

We previously reported apparent ideal diamagnetism at ambient conditions in a graphene-n-heptane-permalloy system. At the same time, the experiments revealed inconsistent behavior, including signal freezing and occasional paramagnetic responses. Further measurements performed without graphene produced similar signals, indicating that graphene is not responsi

Anna Bykhovskaya, Nour Meddahi

This paper presents a framework for binary autoregressive time series in which each observation is a Bernoulli variable whose success probability evolves with past outcomes and probabilities, in the spirit of GARCH-type dynamics, accommodating nonlinearities, network interactions, and cross-sectional dependence in the multivariate case. Existence and uniquen

Xavier Cabre, Giovanni Catino, Luciano Mari, Paolo Mastrolia

We describe a method to prove new integral inequalities for stable minimal hypersurfaces in Euclidean space. As an application, we give a simple proof that complete, two sided, stable minimal hypersurfaces in $\mathbb{R}^4$ are hyperplanes. A core part of the argument hinges on the fact that stable minimal hypersurfaces in non-negatively curved spaces are ex

Xiang Shi, Peng Hu

Climate change has intensified extreme weather and wildfire conditions globally. Canada experienced record-breaking wildfires in 2023 and 2025, burning millions of hectares and severely impacting the Prairie provinces, with Manitoba facing its worst season in 30 years. These events highlight the urgent need to understand and mitigate escalating fire risks. W

Piero Giacomelli

We study log-concavity properties of real sequences $(a_n)_{n \ge 0}$ satisfying a $d$-th order linear recurrence whose coefficients are linear functions of $n$; the so-called P-recursive (or holonomic) sequences. Writing the recurrence in companion-matrix form $\mathbf{v}_{n+1} = M_n\,\mathbf{v}_n$ with $M_n = nA + B$, we show that the log-concave operator

E. S. Andriyakhina, M. Mansouri, M. Breitkreiz, P. W. Brouwer

We present a theory of ballistic N/F/S and S/F/S junctions with a uniformly precessing magnetization, which generates long-range equal-spin superconducting correlations [Takahashi et al., Phys. Rev. Lett. 99, 057003 (2007), Houzet, Phys. Rev. Lett. 101, 057009 (2008)]. The non-equilibrium distribution of Andreev bound states leads to a strongly non-sinusoida

Hyunkyung Park, Arkaitz Zubiaga

Automated fact-checking in dialogue involves multi-turn conversations where colloquial language is frequent yet understudied. To address this gap, we propose a conservative rewrite candidate for each response claim via staged de-colloquialisation, combining lightweight surface normalisation with scoped in-claim coreference resolution. We then introduce BiCon

Sabab Ishraq, Aarushi Aarushi, Juncai Jiang, Chen Chen

Humans routinely infer taste, smell, texture, and even sound from food images a phenomenon well studied in cognitive science. However, prior vision language research on food has focused primarily on recognition tasks such as meal identification, ingredient detection, and nutrition estimation. Image-based prediction of multisensory experience remains largely

Harshit Rajgadia, Jiuci Xu

Recent work has emphasized a subtlety of large- $N$ limits in AdS/CFT: a sequence of pure states in the microscopic theory need not remain pure with respect to the emergent algebra of observables. We study this phenomenon for Kourkoulou-Maldacena (KM) states in the double-scaling limit of the SYK model, and show that their ensemble-averaged algebraic descrip

Dongxin Guo, Jikun Wu, Siu-Ming Yiu

Large Language Model (LLM) agents are increasingly deployed in multi-agent systems requiring strategic coordination. While recent work has analyzed LLM behavior in two-player games, coalition formation, where $n$ agents dynamically form cooperative groups, remains theoretically uncharacterized. We present the first framework grounding coalition formation in

P. Pedroni, F. Afzal, S. Abt, P. Achenbach

New data for the total inclusive helicity-dependent cross section for the proton and deuteron were obtained in the photon energy interval 200-1400 MeV. The experiment was performed at the A2 tagged-photon facility of the Mainz Microtron (MAMI) using a circularly polarized photon beam and longitudinally polarized proton and deuteron targets. The reaction prod

Zengrui Han

We construct minimal resolutions of pushforwards of structure sheaves of toric substacks of smooth toric stacks by line bundles as strong deformation retracts of cellular resolutions constructed by Hanlon, Hicks and Lazarev. We also provide a canonical and combinatorial description of the differentials of such minimal resolutions. Two key ingredients are the

Michael Dougherty, Jon McCammond

This article describes a natural piecewise Euclidean bi-simplicial cell structure for the space of $n$-element multisets in a fixed Euclidean rectangle. In particular, we highlight some connections with spaces of complex polynomials and permutahedra.

Eugenia Pyurbeeva, Ronnie Kosloff

The Lindblad (GKLS) master equation, which represents the mathematical form for the general evolution of a density matrix, is a versatile and widely-used tool in open quantum systems. In contrast with the typical approach of imposing additional conditions on the system, such as weak coupling or energy conservation, we explore the structure of the equation wi

Bao Bach, Cameron Ibrahim, Reuben Tate, Jad Salem

Many combinatorial optimization problems admit a maximin fairness variant, where the aim is to find a distribution over possible solutions which maximizes an expected worst-case outcome. However, the support for an optimal distribution may be exponential, which can be intractable to represent in the worst case. To this end, we propose a quantum based approac

Randeep Ravesh, A R Harikrishnan, Purbarun Dhar

We investigate the thermo solutal transport phenomena and deposition patterns during the evaporation of surfactant laden droplets experimentally and through theoretical scaling based analysis. Experiments were conducted using the sessile droplet configuration in the acrylic chamber for both hydrophilic and hydrophobic substrates. Infrared thermography and pa

Qi Zhang, Dawei Wang, Shaofeng Zou

Reinforcement learning (RL) has emerged as a powerful tool for aligning diffusion models with human preferences, typically by optimizing a single reward function under a KL regularization constraint. In practice, however, human preferences are inherently pluralistic, and aligned models must balance multiple downstream objectives, such as aesthetic quality an

Alan F. Tinoco

The Klein Paradox -- the anomalous scattering of relativistic fermions off a high potential step -- signals the limit of the single-particle interpretation of the Dirac equation. While Quantum Field Theory (QFT) resolves this via pair production, the microscopic mechanism is often obscured by abstract formalism. In this work, we investigate this phenomenon t

M. Szyszko, M. Birowska

Here we analyze multiple symmetry-inequivalent atomic configurations across the entire composition range of the isovalent and isostructural Mo$_x$W$_{1-x}$S$_2$ alloy using density-functional theory and Monte Carlo simulations. Our results show that although structural stability and energetics are largely composition-driven, the electronic and optical proper

Rustem Khasanov, Hubertus Luetkens, Nikolai D. Zhigadlo

We report zero-field (ZF) and transverse-field (TF) muon-spin rotation/relaxation ($\mu$SR) measurements on superconducting Li$_{0.95}$FeAs ($T_{\rm c}\simeq16.0$ K) grown by a high-pressure self-flux method. The ZF-$\mu$SR data show no detectable change of the electronic relaxation rate on cooling through $T_{\rm c}$, providing no evidence for time-reversal

Noureddine Kermiche

Catastrophic forgetting remains a primary hurdle in sequential task learning for artificial neural networks. We propose a silicon-native modular architecture that achieves structural parameter isolation using Task-Specific Experts and a distributed, outlier-based Gatekeeper. Moving beyond traditional sequential consolidation, our framework utilizes a Simulta

Xue Wu, Shengting Cao, Shenglin Li, Jiaqi Gong

Rural environmental risks are shaped by place-based conditions (e.g., housing quality, road access, land-surface patterns), yet standard vulnerability indices are coarse and provide limited insight into risk contexts. We propose SatBLIP, a satellite-specific vision-language framework for rural context understanding and feature identification that predicts co

Aaron Jones, Marija Ilic

This paper presents a comparative AC Optimal Power Flow study on a real world city scale islanded microgrid with high solar PV penetration, implemented within a Digital Twin framework. Four objective function cases economic dispatch, voltage stress exposure via PV power factor variation, then optimal load delivery, and capacitor enhanced economic dispatch as

Christiane Ernst, Luis Gutmann, Domenique Zipperling, Kathrin Figl

In high-stakes AI-supported decisions, considerations are not purely technical but involve moral judgments about fairness, responsibility, and harm. While prior research has focused mainly on functional or behavioral alignment, this paper argues that moral alignment may be a more fundamental dimension of human-AI decision-making. Moral alignment is defined a

Carri W. Chan, Yi Han, Hannah Li, Benjamin L. Ranard

AI tools increasingly guide targeted interventions in healthcare, education, and recruiting. Algorithms score individuals, trigger outreach to those above a threshold (e.g., high-risk or high-value), and encourage them to request service; then providers deliver service to those who request. Standard practice sets the threshold and selects the algorithm to ma

Nemin Wei, Felix von Oppen, Leonid I. Glazman

Twisted bilayer graphene provides a paradigmatic platform for exploring the interplay between electronic topology and strong correlations. Within the topological heavy fermion model [Song and Bernevig, Phys. Rev. Lett. 129, 047601 (2022)], topology and electron interactions are brought together by including a weak hybridization between the bands of itinerant

Figen Oztoprak, Richard Byrd

We propose a sequential quadratic programming (SQP) algorithm for inequality constrained optimization that is robust to the presence of bounded noise in function and derivative evaluations. We cover the case where constraint evaluations contain noise as well as the objective. The proposed algorithm is a line search SQP method with relaxations to deal with no

Deniz Coskun, R. Chitra

We develop a general framework to calculate the many-body density of states (DOS) of isolated and interacting quantum systems. Based on the generalized coherent state formalism and the Simon-Lieb bounds for a quantum partition function, our method provides a general method of calculation for the DOS in high-dimensional irreducible sectors. This framework fur

José N. V. Gomes, Willian I. Tokura, Hikaru Yamamoto

We study the Ricci-Bourguignon flow on warped product manifolds with noncompact base. This setting leads naturally to a parabolic partial differential equation on the space of smooth warping functions, arising from the necessary and sufficient conditions for a warped metric to evolve under the flow. One of our main results establishes a gradient estimate for

Nguyen Phan, Brian Kim, Adeel Zafar, Guoning Chen

Streamlines have been widely used to represent and analyze various steady vector fields. To sufficiently represent important features in complex vector fields (like flow), a large number of streamlines are required. Due to the lack of a rigorous definition of features or patterns in streamlines, user interaction and exploration are required to achieve effect

Julien Martinelli, Ibtissem Rebai, David W. Haas, Julie Bertrand

High-dimensional genetic covariate selection in population pharmacokinetic (PK) models is challenging due to the cohort's restricted size and high correlation among single-nucleotide polymorphisms (SNPs). We propose a fully Bayesian, single-stage framework that jointly infers nonlinear mixed effect model (NLMEM) parameters and SNP effect sizes, providing coh

Akshay Paruchuri, Ishan Chatterjee, Henry Fuchs, Ehsan Adeli

Multimodal language models systematically underperform on visual perception tasks, yet the structure underlying this failure remains poorly understood. We propose centroid replacement, mapping tokens to their nearest K-means centroid and removing within-cluster residual structure, as a controlled probe for modal dependence. Across seven models spanning four

Prithwish Basu Roy, Zeng Wang, Anatolii Chuvashlov, Weihua Xiao

Large Language Models (LLMs) have shown significant improvement in RTL code generation. Despite the advances, the generated code is often riddled with common vulnerabilities and weaknesses (CWEs) that can slip by untrained eyes. Attackers can often exploit these weaknesses to fulfill their nefarious motives. Existing RTL bug-detection techniques rely on rule

Pratyay Banerjee, Masud Moshtaghi, Shivashankar Subramanian, Amita Misra

Large language models still struggle with reliable long-term conversational memory: simply enlarging context windows or applying naive retrieval often introduces noise and destabilizes responses. We present APEX-MEM, a conversational memory system that combines three key innovations: (1) a property graph which uses domain-agnostic ontology to structure conve

Saurabh Bagchi, Hyunseung Kim, Tarek Abdelzaher, Homa Alemzadeh

Resilience in cyber-physical systems (CPS) is the fundamental ability to maintain safety and critical functionality despite adverse "perturbations," which includes security attacks, environmental disruptions, and hardware or software failures. This survey provides a comprehensive review of CPS resilience, framing the field through five interconnected themes

Danila Astrakhantsev, Sebastiaan Krijt, Sofia Savvidou, Bertram Bitsch

Pebble drift plays a central role in modern planet formation models. In this work we carry out planet formation simulations (including pebble accretion and migration) for a range of disc parameters to investigate (a) the impact of the snowline pebble mass flux on final planet orbits and masses, and (b) the back-reaction of growing and migrating planets on th

Jeffrey C. Ching, Quan Zhou, Danfeng Zhang

Existing language-based information-flow control (IFC) tools face a fundamental tension: Denning-style systems that track explicit and implicit flows at the variable level typically require compiler modifications, while more coarse-grained approaches, including recent work Cocoon, avoid compiler changes but impose more restrictive programming models. We pres

Ravi Kini, David Doty

Chemical reaction networks, or CRNs, are known to stably compute semilinear Boolean-valued predicates and functions, provided that all reactions are irreversible. However, this property does not hold for wet-lab implementations, as all chemical reactions are reversible, even at very slow rates. We study the computational power of CRNs under the reverse-robus

Hsiang-Chen Yeh, Luqi Sun, Aurosweta Mahapatra, Shreeram Suresh Chandra

This study investigates whether speech-based depression detection models learn depression-related acoustic biomarkers or instead rely on speaker identity cues. Using the DAIC-WOZ dataset, we propose a data-splitting strategy that controls speaker overlap between training and test sets while keeping the training size constant, and evaluate three models of var

Qiyang Lyu, Zhenyu Wu, Wei Wang, Hongming Shen

Localization of autonomous mobile robots (AMRs) in enclosed or semi-enclosed environments such as offices, hotels, hospitals, indoor parking facilities, and underground spaces where GPS signals are weak or unavailable remains a major obstacle to the deployment of fully autonomous systems. Infrastructure-based localization approaches, such as QR codes and RFI

Avinash Amudala

Online A/B testing at scale relies on proxy metrics -- short-term, easily-measured signals used in place of slow-moving long-term outcomes. When the proxy-outcome relationship is heterogeneous across user segments, aggregate correlation can mask directional failures akin to Simpson's Paradox, leading to costly ship/no-ship errors. We introduce PROXIMA (Proxy

Michael J. O'Neill, Aoji Tang

In this paper, we consider nonlinear optimization problems with a stochastic objective function and deterministic equality constraints. We propose an inexact two-stepsize stochastic sequential quadratic programming (SQP) algorithm and analyze its worst-case complexity under mild assumptions. The method utilizes a step decomposition strategy and handles stoch

William Bennett, Ryan G. McClarren, Ethan Smith, Melek Derman

Dynamic Mode Decomposition (DMD) is a data-driven method for approximating the spatiotemporal modes of a system. The eigenvectors and eigenvalues of the system are approximated from a series of time-snapshots of the state variables. The standard formulation of DMD is subject to strict assumptions concerning the time-spacing of the snapshots and is biased by

Ting Wang, I. Brian Becerril-Castro, Ana Sousa-Castillo, Miguel A. Correa-Duarte

Surface-enhanced Raman scattering (SERS) combines analyte-specificity and single-molecule sensitivity, but its potential is limited by slow readout where sophisticated nanosensors are analysed in a serial fashion, one particle at a time. We introduce SERS lock-in sampling to resolve the decades-old trade-off between spectral resolution and widefield imaging.

Qihui Bu, Yiqiang Q. Zhao

In 1985, Grassmann, Taksar, and Heyman published their celebrated paper, in which they introduced a numerically stable algorithm for computing the stationary probabilities of a finite-state Markov chain, one of the key performance quantities in both theory and applications. This algorithm later became the well-known GTH algorithm (or the state-reduction meth

Tianhao Qian, Jiayu Chen, Lixu Wang

LLM-guided search is usually adopted to solve complex tasks by ranking and pruning top-$K$ candidates based on evaluator scores. However, irreducible bias still exists even if popular methods, such as repeated sampling, are applied to reduce variance. Consequently, pruning may remove every continuation that can reach a valid solution. In this paper, we propo

Daniëlle L. A. Pieterse, Andrew J. Levan, Maria E. Ravasio, Jillian C. Rastinejad

We present detailed, multi-wavelength analysis of GRB 210704A: a Fermi Gamma-ray Burst Monitor discovered and Fermi Large Area Telescope (LAT) detected gamma-ray burst (GRB). The burst is dominated by a short ($\approx 2$ s) pulse followed by weaker, softer emission. We line stack our afterglow spectrum and determine the most likely redshift to be $z = 2.34$

Siddharth Bhatnagar, Francis Codron, Ehouarn Millour, Emeline Bolmont

Ocean dynamics are often sidelined in exoplanet climate studies due to the high computational cost of fully coupled atmosphere-ocean general circulation models (GCMs). However, ocean heat transport (OHT) can play a critical role in shaping the climate and observables of terrestrial planets. As a compromise, most exoplanet GCMs rely on slab ocean models witho

Shao-Yu Fu, Cui-Yuan Dai, Ai-Ling Wang, Dong Xu

Extragalactic Fast X-ray Transients (EFXTs) represent an emerging class of high-energy phenomena characterized by X-ray outbursts lasting from tens to hundreds of seconds. However, for more than half of the EFXTs, their physical origins remain elusive. In this Letter, we report the discovery of EP250302a, a luminous EFXT detected by the Einstein Probe (EP) a

Zichong Li, Chen Liang, Liliang Ren, Tuo Zhao

Large language models (LLMs) increasingly operate in settings that require reliable long-context understanding, such as retrieval-augmented generation and multi-document reasoning. A common strategy is to fine-tune pretrained short-context models at the target sequence length. However, we find that standard long-context adaptation can remain brittle: model a

Firuz Kamalov, Fadi Thabtah, R. Sivaraj, Neda Abdelhamid

We introduce path-sampled integrated gradients (PS-IG), a framework that generalizes feature attribution by computing the expected value over baselines sampled along the linear interpolation path. We prove that PS-IG is mathematically equivalent to path-weighted integrated gradients, provided the weighting function matches the cumulative distribution functio

Lara Daw, Najmeddine Attia

$\Phi$-intermediate dimensions interpolate between Hausdorff and box-counting dimensions by restricting admissible coverings to scale windows of the form $[\Phi(r),r]$. Using a family of $\Phi$-dependent kernels, we develop a potential-theoretic framework that characterizes these dimensions in terms of capacities and leads to associated $\Phi$-dimension prof

Pushpa Kumar Balan, Aijing Feng

Gradient saliency from deep sequence models surfaces candidate biomarkers efficiently, but the resulting gene lists can be contaminated by tissue-composition confounders that degrade downstream classifiers. We study whether LLM chain-of-thought (CoT) reasoning can filter these confounders, and whether reasoning quality is associated with downstream performan

Yuncong Liu, Yuan Wan, Zhou Jiang, Yao Lu

Key Opinion Leader (KOL) discourse on social media is widely consumed as investment guidance, yet turning it into executable trading strategies without injecting assumptions about unspecified execution decisions remains an open problem. We observe that the gaps in KOL statements are not random deficiencies but a structured separation: KOLs express directiona

Aditi De

Diffusion-model inference and overdamped Langevin dynamics are formally identical. A physical substrate that encodes the score function therefore equilibrates to the correct output by thermodynamics alone, requiring no digital arithmetic during inference and potentially achieving a $10{,}000\times$ reduction in energy relative to a GPU. Two fundamental barri

Dmitry Eremeev, Salem Said, Viacheslav Borovitskiy

Applying kernel methods to matchings is challenging due to their discrete, non-Euclidean nature. In this paper, we develop a principled framework for constructing geometric kernels that respect the natural geometry of the space of matchings. To this end, we first provide a complete characterization of stationary kernels, i.e. kernels that respect the inheren

Abubakar Sadiq Shittu, Clay Shubert, John Sadik, Scott Ruoti

"Cryptography turns a security problem into a key management problem." Despite decades of research effort towards usable key management, it remains unclear whether key management issues are inherent to every cryptographic system or merely artifacts of specific designs. To investigate, this paper presents a user study of mutual TLS (mTLS) usability, t

Bryan Jhoan Cazáres Leyva, Ulises Gachuz Davila, José Juan González Fonseca, Juan Irving Vasquez

Non-violent street robberies (snatch-and-run) are difficult to detect automatically because they are brief, subtle, and often indistinguishable from benign human interactions in unconstrained surveillance footage. This paper presents a hybrid, pose-driven approach for detecting snatch-and-run events that combines real-time perception with an interpretable cl

Patricio A. Gallardo, Kris Pardo, Oliver H. E. Philcox, Nicholas Battaglia

The mean pairwise velocity of massive halos reflects the gravitational force law on cosmic scales. We combine cosmic microwave background intensity maps from the Atacama Cosmology Telescope and a galaxy catalog from the Sloan Digital Sky Survey to estimate the mean pairwise velocity using the kinematic Sunyaev-Zeldovich (kSZ) effect. On scales from 30 -- 230

Dmitriy Bilyk, Liudmyla Kryvonos, Ryan W. Matzke, Edward Saff

We investigate the asymptotic behavior of greedy $s$-Riesz and Green energy sequences $\{x_{n}\}_{n=1}^{\infty}$ on the unit sphere $\mathbb{S}^{d} \subset \mathbb{R}^{d+1}$, where each point $x_n$ is defined as the minimizer of the discrete potential generated by the preceding points $x_1, x_2, ..., x_{n-1}$. We show that the greedy sequence attains optimal

Bar Alon, Itamar Zimerman, Lior Wolf

Large language models (LLMs) achieve strong performance and have revolutionized NLP, but their lack of explainability keeps them treated as black boxes, limiting their use in domains that demand transparency and trust. A promising direction to address this issue is post-hoc text-based explanations, which aim to explain model decisions in natural language. Pr

Hao An, Yibin Lou, Jiayi Guo, Yang Xu

Large language models (LLMs) often exhibit hallucinations due to their inability to accurately perceive their own knowledge boundaries. Existing abstention fine-tuning methods typically partition datasets directly based on response accuracy, causing models to suffer from severe label noise near the decision boundaries and consequently exhibit high rates of a

Hela Mhiri, Hugo Thomas, Léo Monbroussou, Ulysse Chabaud

Boson sampling is a leading candidate for demonstrating quantum advantage in photonic systems. Despite significant experimental and theoretical progress, a characterization of its output statistics remains incomplete. This is especially true beyond the dilute regime, where photon collisions and bunching become significant. The associated saturated regime, ch

Chandrasekhar Gokavarapu

Let \(R\) be a commutative ring and \(M\) an \(R\)-module. We develop a localization and local-global theory for \(C4\)-modules, \(C4^{\ast}\)-modules, strongly \(C4^{\ast}\)-modules, \(C4\)-hulls, and pseudo-continuous hulls over commutative rings. The problem is structural: these notions are defined through decompositions, summand conditions, and minimal e

Horace Yiu, Leandro Sánchez-Betancourt, Álvaro Cartea, Gerardo Duran-Martin

We derive a robust update rule for the online infinite hidden Markov model (iHMM) for when the streaming data contains outliers and the model is misspecified. Leveraging recent advances in generalised Bayesian inference, we define robustness via the posterior influence function (PIF), and provide conditions under which the online iHMM has bounded PIF. Imposi

Jason Potteiger, Andrew Hong, Ito Zapata

We tasked GPT-4.1 to read what baseball fans wrote about their game-day experience and predict the overall experience rating each fan gave on a 0-10 survey scale. The model received only the text of a single open-ended response. These AI predictions were compared with the actual experience ratings captured by the survey instrument across approximately 10,000

J. M. Lindert, L. Mai

We present the implementation of next-to-next-to-leading order (NNLO) electroweak (EW) virtual corrections at next-to-leading logarithmic (NLL) accuracy in the amplitude generator OpenLoops. The implementation covers the automated computation of processes involving massless fermions and transversely polarised vector bosons. For energies above the EW scale, l

Haomin Zhuang, Hanwen Xing, Xiangliang Zhang

Recent autonomous AI agents such as Codex, and Claude Code have made it increasingly practical for users to delegate complex tasks, including writing emails, executing code, issuing shell commands, and carrying out multi-step plans. However, despite these capabilities, human-agent interaction still largely happens through terminal interfaces or remote text-b

Enrico Bozzetto, Jonte R. Hance

In the literature, there are two differing definitions of contextuality: Kochen and Specker's, and Spekkens' (or ``generalised''). However, researchers using one of these definitions rarely consider the other, meaning comparative analysis of these two notions is rare. In this paper, we advance the idea that Kochen-Specker contextuality provid

Wolfgang König

We consider the interacting Bose gas in the thermodynamic limit in a large box in $\R^d$ at positive temperature $1/\beta\in(0,\infty)$ with particle density $\sim\rho\in(0,\infty)$. We follow a path-integral approach and adopt from \cite {ACK10} a description of the free energy in terms of the {\it Brownian loop soup}, a Poisson point process consisting of

Salem Radey, Jack West, Kassem Fawaz

Agentic systems built on large language models (LLMs) are increasingly being used for complex security tasks, including binary reverse engineering (RE). Despite recent growth in popularity and capability, these systems continue to face limitations in realistic settings. Cutting-edge systems still fail in complex RE scenarios that involve obfuscation, timing,

Kinhei Lee, Peiyuan Jing, Zhenxuan Zhang, Yue Yang

Large scale vision language models have shown promise in automating chest Xray interpretation, yet their clinical utility remains limited by a gap between model outputs and radiologist reasoning. Most systems optimize for semantic information without emulating how experts visually examine medical images, often overlooking critical findings or diverging from

Emily Lugos, Maurício Gruppi

The modern news cycle has been fundamentally reshaped by the rapid exchange of information online. As a result, media framing shifts dynamically as new information, political responses, and social reactions emerge. Understanding how these narratives form, propagate, and evolve is essential for interpreting public discourse during moments of crisis. In this s

Gabriel Pimenta de Freitas Cardoso, Caio Lucas da Silva Chacon, Jonas Felipe da Fonseca Oliveira, Paulo Henrique de Medeiros Araujo

This manuscript introduces DharmaOCR Full and Lite, a pair of specialized small language models (SSLMs) for structured OCR that jointly optimize transcription quality, generation stability, and inference cost. It also presents DharmaOCR-Benchmark, a benchmark that covers printed, handwritten, and legal/administrative documents, and proposes a unified evaluat

Kuan Wang, Julian B. Muñoz, L. Y. Aaron Yung

High- to ultrahigh-redshift clustering of halos provides a powerful tool to understand cosmology and galaxy formation. However, theoretical predictions are not firmly established in the first billion years, where current and upcoming surveys are beginning to reach percent-level precision. Here we measure dark matter halo biases at $z=6$ - 19 from simulation

Hao Ran, Daniel Verscharen, Jesse Coburn, Georgios Nicolaou

Observations of solar wind velocity distribution functions (VDFs) commonly reveal fine-scale structures. These features strongly influence kinetic processes such as wave damping and instability, yet their role remains poorly understood. We use a Gaussian Mixture Model (GMM) to separate proton and $\alpha$-particle (fully ionized helium) VDFs from Solar Orbit

Li-Hsiang Shen, Yu-Quan Zheng

To address high data traffic demands of sixth-generation (6G) networks, this paper proposes a novel architecture that integrates autonomous aerial vehicles (AAVs) and multi-functional reconfigurable intelligent surfaces (MF-RISs) as AM-RIS in fluid antenna (FA)-assisted full-duplex (FD) networks. The AM-RIS provides hybrid functionalities, including signal r

Mohammad Mirtaba, Max H. Cohen

The combination of control barrier functions (CBFs) and adaptive control -- a framework referred to as adaptive safety -- has proven to be a powerful paradigm for safety-critical control of nonlinear systems with parametric uncertainties. Yet the theoretical conditions for forward invariance within this framework are often quite conservative, and may require

Michaela Bush, Jeremy C. Palmer, Michael P. Howard

We develop a method for simulating colloidal suspensions using multiparticle collision dynamics (MPCD) with a discrete particle model represented as a rigid body. The key steps for incorporating the rigid-body constraints are to thermalize the velocities of the discrete sites before they participate in the MPCD collision step, then transfer momentum from the