October 2026 arXiv papers — page 3

Gabriele Nani, Simone Vincini

In general, the capillarity energy outside an unbounded obstacle may fail to admit volume-constrained minimizers, as infimizing sequences may escape to infinity. We study this minimization problem outside a closed convex obstacle $C$. Via a $Γ$-convergence result for the capillarity energies outside sequences of closed convex obstacles converging in the Haus

Ziyang Song, Lia Shen, Yixuan Li, Qincheng Lu

Biomedical language models (LMs) provide strong representations for clinical tasks, but they often fail to capture the hierarchical structure among medical concepts. This limitation is especially pronounced for rare and fine-grained clinical codes in long-tailed vocabularies, where hierarchy-agnostic LM representations become less informative as concept freq

Chengchao Yuan

High-energy neutrinos from Seyfert galaxies probe particle acceleration near supermassive black holes, but the accelerated nuclear composition remains uncertain. We connect stochastic acceleration to nuclear and radiative cascades to assess nuclear contributions. Acceleration and losses limit the energy per nucleon available to nuclear cascades; without seco

Rohit Mohan

Reliable robot autonomy requires reasoning about the world from partial and noisy observations. However, many perception systems model only visible scene regions, rely on fixed semantic categories, and assume reliable sensor measurements. These assumptions break under occlusion, novel object encounters, and sensor degradation. In this paper, I present a rese

Johanna Magdalena Kuch, Maud van Lier, Daksitha Senel Withanage Don, Patrick Gebhard

Recent advances in zero-shot voice cloning systems make it possible to reproduce a recognizable voice from a short recording and attach it to arbitrary virtual characters. Prior work suggests that familiar vocal identity can carry over associated personality impressions to new targets such as robots or virtual characters. However, these works have mainly exa

Chichen Lin, Huaijin Xie, Wenhan Xie, Weijian Fan

Crowdsourced misinformation governance systems such as Community Notes and Weibo Annotations depend on sustained corrective labor. How monetary incentives shape participation and subsequent governance outcomes remains poorly understood. We conduct a mixed-methods study of Weibo Annotations, analyzing 102,994 posts and 431,407 comments from 2023--2026, alongs

Wen Si

For $Q_α(z)=e^{2πiα}z+z^2$ with $α$ irrational, we prove that the analytic centraliser of $Q_α$ is trivial whenever $Q_α$ is not analytically linearizable. Together with the quadratic Brjuno--Yoccoz theorem this gives a complete dichotomy, for the centraliser is infinite cyclic at the non-Brjuno parameters and conjugate to the group of nonzero complex dilati

Liuquan Wang

We prove four conjectures of Takigiku and Tsuchioka consisting of eight Rogers--Ramanujan type identities. Two conjectures involve triple sums and a third involves quadruple sums, and their product sides are certain level $2$ principal characters of $A_{13}^{(2)}$. The fourth conjecture gives multisum representations of certain level $2$ principal characters

Simone Grillo, Offir Zachs Maishlos, Liya Koren, Rotem Strassberg

Double perovskites (DPs) offer a compositionally versatile platform for photocatalysis and optoelectronics, but their soft lattices favor strong exciton-phonon coupling and self-trapped excitons (STEs), which produce broad, red-shifted photoluminescence (PL). Here, we combine optical spectroscopy and ab initio calculations to study lead-free, Sb-doped Cs2BIn

Jiusong Ge, Mingzhao He, Qidong Liu, Chunbao Wang

A patient's pathology case spans tens of slides from distinct specimens and anatomical sites, each carrying non-redundant, site-specific evidence. Forming a patient-level conclusion therefore requires locating the slides relevant to a given question and integrating complementary findings across them. However, existing pathology agents analyze one slide a

S. N. Artekha

A possible mechanism of tornado generation is associated with the Schwarzschild criterion for convective instability. The supercooled cloud-air mixture near the axis starts to descend and, due to the super adiabatic temperature gradient, the overheated near-surface air mass rises. The complete system of equations for full model includes equations for the str

Jakob Loren

Let $n\ge5$ be odd and let \(L_{3q}^n=S^n/C_{3q}\) be the standard lens space, where $q>3$ is prime. We prove that for all sufficiently large $q$, \[ σ(L_{3q}^n) > (3q)^{-2/n}Y(S^n)=Y(L_{3q}^n,[g_{\rm rd}]). \] Here \([g_{\rm rd}]\) denotes the quotient-round conformal class. Combining spin bordism with the surgery theorem of Ammann--Dahl--Humbert, we establ

Maria Pfahler, Andrea Donarini

We propose a protocol to prepare and probe dark states in a single-molecule electron spin resonance based on atomic force microscopy (ESR-AFM), thus extending existing techniques developed to address Rabi oscillations. To this end, we simulate in analogy to atomic physics the coherent manipulation of a three-level system in the V-configuration using a multif

Karthikeya S M Yelisetty, Shamik Sural, Santosh Nagarakatte

This paper proposes a methodology for decoupling the authorization logic of a relational database from its core data management engine, enabling specification of advanced access control policies with minimal changes to the code base. Our decoupled authorization architecture, which we call ARCADE, externalizes access request resolution to a formally verifiabl

Lee A. Whitaker, Michael Tremmel, Angelo Ricarte, Priyamvada Natarajan

Massive black holes (MBHs) grow through both gas accretion and mergers and it remains unclear how variations to MBH growth and dynamical histories contribute to the spread in empirical scaling relations between MBHs and their host galaxies. We examine the ex- and in-situ growth of MBHs within large-scale cosmological simulations and compare the evolution of

Mengyu Li, Qianqian Qu, Jun S. Liu

Generative models are often required to serve many downstream conditional tasks. In the training-free setting, a pretrained unconditional model is reused as a prior and combined with a task-specific likelihood to sample from the corresponding posterior. While such training-free samplers have been developed for diffusion and flow models, their counterpart for

Nikos Frantzikinakis, Davide Sclosa

We develop a new approach to the Joint Intersectivity Conjecture of Bergelson, Leibman, and Lesigne for commuting transformations. The conjecture asserts that the necessary divisibility condition is the only obstruction to polynomial multiple recurrence and, consequently, to the multidimensional polynomial Szemerédi theorem. We establish the conjecture for p

Weida Liang, Kenji Kawaguchi

Language model systems repeatedly make test-time decisions by comparing similarity: they retrieve exemplars, rank candidates, and revise responses. Yet similarity is not unique: lexical, semantic, structural, and stylistic resemblance can point to different candidates, and the task-aligned notion can change across instances. We introduce BASIS (Bayesian Axis

Ido Madar, Gal Oren, Shai Avidan

Texture interpolation asks what should lie between two textures. Standard methods typically blend pixels or features at corresponding spatial locations, implicitly assuming that co-located elements should be mixed. For unrelated textures, however, this correspondence is arbitrary: a crack at one location has no particular relationship to a fiber at the same

OneRec Team, Chenglong Chu, Guohong Mu, Guowang Zhang

Summary-token architectures such as Kwai Summary Attention (KSA) compress distant context into learnable Summary token per chunk, and hybrid variants interleave these layers with full attention to retain exact access to the original text. In such hybrid KSA-Full architecture, the interleaving inserted summary tokens are served as a streaming flow role across

Ying Jiang, Ivan Saika-Voivod

Rare events, such as crystal nucleation in mildly supercooled liquids, are characterized by waiting times that can be much longer than the timescale of the underlying microscopic dynamics. The mean first-passage time (MFPT) formalism [J. Wedekind and D. Reguera, J. Phys. Chem. B 112, 11060 (2008)] provides a route to reconstruct one-dimensional free-energy p

Peiwei Chang, Li Zhang, Mark D. Plumbley, Philip J. B. Jackson

Audio memory must recognise recurring sources across acoustic changes while keeping their attributes up to date. Overall accuracy does not reveal how these demands interact. We introduce Audio Memory Operations (AMemOps), a benchmark for latest-attribute retrieval, preservation of unchanged attributes and absent-source rejection in long audio streams. Paired

Leif Sieben, Yoel Zimmermann

Chemical language models (CLMs) are pretrained on large corpora of molecules with the aim of learning representations that transfer to diverse downstream tasks. Here, we systematically study the effects of pretraining using a series of such CLMs across 64 molecular property prediction datasets. We find that the downstream gains of pretraining in CLMs do not

Chenru Wen, Huajian Zeng, Haokai Ding, Jian Kang

Vision-language-action (VLA) policies perform manipulation tasks using camera images and language instructions, but object slip or visual occlusion can disrupt execution. Recovery is typically learned during policy training or delegated to a monitor that triggers a scripted repair after detecting a failure. We introduce ForkCast, a predictive recovery framew

Tommaso Salvatori, Matteo Forasassi, Alexis Toumi, Giovanni de Felice

Message passing neural networks have shown remarkable performance in algorithmic and constraint reasoning, due to their ability to encode the structure of the task directly in the architecture. Recent work has shown that better results can be achieved using recursive transformers trained with deep supervision and truncated recursions, which make extremely de

E. Vardoulaki, T. Shimwell, M. J. Hardcastle, L. Doukas

We present the first deep, targeted LOFAR 144 MHz source catalogue of the COSMOS field, together with the first 144 MHz source counts in this field. The data comprise 48 h of observations. Over the 2.6 sq. deg. region adopted for the catalogue analysis, the final Dutch-array image reaches a median rms sensitivity of ~100 uJy/beam at an angular resolution of

Christoph Widder

We prove the existence of unique solutions ${u\in H^1(0,T;L^2_γ(\mathbb{R}^n))\cap L^2(0,T;H^1_γ(\mathbb{R}^n))}$ of the transport equation $u_t=F\cdot\nabla u+f$ and the orthogonal dynamics equation ${u_t= Q_\text{Z} (F\cdot\nabla) Q_\text{Z} u+f}$, where $F$ is a time-independent vector field, $f$ is a forcing term, $P_\text{Z}=1-Q_\text{Z}$ is a Zwanzig p

Ryunosuke Miyazaki

We construct confidence sets for the entire Fréchet mean set by inverting studentized pairwise risk inequalities. Gaussian multiplier calibration localized around empirical near minimizers gives asymptotic coverage at least the nominal level, uniformly over distribution classes satisfying empirical-process conditions. The construction requires neither unique

Augusta van Haren, Hanna Hoogen, Luca Pattavina, Quincy Rommen

Video prediction requires models to capture complex spatiotemporal dependencies and to retain relevant information over long sequences. PredRNN and its successor PredRNN-V2 address this challenge with spatiotemporal Long Short-Term Memory (ST-LSTM) units that maintain two separate memory states, at the cost of a relatively complex recurrent building block. I

A. Núñez, M. A. Agüeros, J. L. Curtis, C. A. Ospinal Achicanoy

We present a study of the rotation and the chromospheric H$α$ activity of low-mass stars in the $\approx$120 Myr-old Pleiades and the $\approx$400 Myr-old Group X open clusters. We combine published and new rotation periods ($P_\mathrm{rot}$) from TESS and ZTF, and use LAMOST, SDSS, and MDM spectra to measure H$α$ equivalent widths (EWs). While overall the G

Alessio Marrani, Daniele Corradetti

The Okubo algebra brings together ternary gradings, integral root frames and rank-three geometry. We examine how these structures extend to the complex Exceptional Periodicity algebras. With $N=4(n+1)$, a tetracode character gives eight nonzero homogeneous components of dimension $t_n=8n+3+2^{4n}$. On a tip, a compatible grading has zero component of dimensi

Yangchengyu Zhou, Zhiping Wu, Runchang Jiang, Dongdong Ren

The Key-Value (KV) cache constitutes a dominant source of memory consumption during large language model (LLM) inference, particularly in long-context and complex reasoning scenarios. Existing KV cache quantization methods have explored different compression granularities, including layer-wise precision adjustment and token-level information preservation. In

Dhairya Shah, Rajashekar Reddy Chinthalapani, Pramuka Sooriya Patabandige, Ambuj Varshney

Wireless communication remains the most power-consuming operation in embedded systems. Low-power transmitters such as backscatter achieve microwatt-scale operation but produce weak signals susceptible to in-band interference and frequency-selective fading, particularly in non-line-of-sight settings. Multi-band transmission can mitigate these effects through

Simone Travagliati, Maria Luisa Merani

In Intelligent Transport Systems, protecting vulnerable road users, specifically pedestrians, is a paramount challenge. The present work contributes to this field, putting forth a distributed architecture in which vehicles connected to the cellular infrastructure collect pedestrians' trajectories and fine-tune a foundation model for pedestrian trajectory

Heng Guo, Daniel A. Spielman, Xinyuan Zhang

We show that the guaranties of the Gram-Schmidt Walk Algorithm of Bansal, Dadush, Garg, and Lovett can be achieved by an Ising model. Given vectors $v_1, \ldots, v_n$ of norm less than 1, we give a closed form for a centrally symmetric distribution on a vector $σ\in \{\pm 1 \}^n$ so that the signed sum $\sum_j σ(j) v_j$ is $1$-subgaussian. We show that the G

Fabian Gröger, Petar Damjanović, Maria Brbić

Combining pretrained models is a common strategy for building AI systems, whether by aligning a vision encoder with a language model or merging fine-tuned checkpoints. However, finding compatible pairs typically requires costly empirical search, as compatibility is known only after alignment or merging. We propose COMPASS (COMPatibility ASsessment Score), a

Justin McMillen, Yasin Yilmaz

A robot's world model must remember objects that leave the camera's view. FloWM, a learned world model, keeps them in a spatial grid of features that it moves with five built-in velocities. A node memory instead keeps each object's position and velocity as numbers. We compare them in FloWM's own block world, with an oracle-fed Kalman tracker

Irena M. Jovanović, Jelena Sedlar, Riste Škrekovski

For a graph G of order n with adjacency eigenvalues lambda_1(G) >= ... >= lambda_n(G), the cospectrality of G within the set G of all graphs of order n is the smallest spectral distance from G to another graph of G. Abdollahi and Zakeri conjectured that for every n >= 4 the cospectrality of the complete tripartite graph K_{n,n,n}, both in the Manhattan and i

Kérian Fiter, Angelica Portocarrero, Bentley Oakes

The development of a digital twin brings together many stakeholders and technical components. We have previously investigated the reporting of the conceptual architecture for these digital twins as an artifact for discussion. However, we find that the activity of modeling the digital twin for visualization and consensus-building can be quite tedious and time

Jianzhu Bao, Jia-Chen Gu, Qingyuan Liu, Haozhen Zhang

Recent studies investigate the weight geometry through its spectral structure, which characterizes how the weights transform different input directions. Within this view, supervised fine-tuning (SFT) is shown to induce dense updates that perturb the spectral geometry of pre-SFT weights along their principal directions, potentially disrupting the model's

E. Concepción, T. Petrushevska, G. Leloudas, D. Ilić

Transients in galactic nuclei arise from distinct processes, only some of which trace the supermassive black hole (SMBH). These include tidal disruption events (TDEs), changing-look active galactic nuclei, Bowen fluorescence flares (BFFs) and ambiguous nuclear transients, alongside supernovae that appear nuclear but are unrelated to the SMBH. Reliable classi

Bruno Bauwens, German Kuznetsov

Solomonoff induction is a general solution for sequence prediction that started the field of algorithmic information theory around 1960. Let $M$ be the universal semimeasure. After observing a string $ω_1 \cdots ω_{n-1}$, it predicts the next bit $ω_n$ to be $b$ with conditional probability proportional to $M(ω_1 \cdots ω_{n-1} b)$. Thus the error of this pr

Guy Ohad, Nadine C. Bradbury, Joseph E. Subotnik

We demonstrate how to build a robust phase space electronic Hamiltonian $\hat{H}_\text{PS}(\boldsymbol{R},\boldsymbol{P})$ that depends on nuclear position $\boldsymbol{R}$ and nuclear momentum $\boldsymbol{P}$ in the presence of ionic pseudopotentials. In particular, we show that in order to account for the non-local component of a pseudopotential while als

Ying Wang, Yuchao Li, Håkan Hjalmarsson

We consider an infinite-horizon discounted dynamic programming (DP) problem for a countable-innovation belief-state model. The model is a surrogate for a linear-Gaussian dual control problem involving an unknown parameter vector and Gaussian observation noise. The surrogate preserves the first two moments of the Gaussian innovation and the resulting posterio

Ipsita Mandal

We study collective charge excitations in pseudospin-1 triple-point semimetals within the random-phase approximation, considering both the linearised model with a flat-band and its quadratic-corrected counterpart. We analyse their density response, Drude weights, screening, plasmon dispersion and damping, particle-hole continua, and Friedel oscillations from

Colin Wilkins, Ryan Caron, Ethan Tsai, Chris Shaffer

We present the design methodology, implementation, and data products of the Energetic Particle Detector (EPD) instrument on-board the Electron Losses and Fields Investigation (ELFIN) dual SmallSat mission. ELFIN was designed to study the time-evolution of energetic particle distributions in Earth's magnetosphere, having operated in polar Low Earth Orbit

Zonghuan Li, Chirag Raman, Hayley Hung

Interactions happen universally and spontaneously in daily lives. Studying involvement in interactions helps us analyze who can influence who and has a wide potential in applications. Specifically, we consider conversational involvement in free-standing conversations. We study two operationalizations of an FCG: F-formations and conversation floors, and propo

Olivier Bénichou, Aurélien Grabsch

The symmetric exclusion process (SEP) is a paradigmatic model of interacting particle transport. Its current--density correlations are governed by an infinite many-body hierarchy of equations. Only recently, this hierarchy was found to collapse at large scales onto a single closed equation. Is this remarkable simplification created by coarse graining, or doe

Eduardo Perez-Richet, Vasiliki Skreta

Relying on manipulable characteristics to target the allocation of resources is fraught with difficulties. Naively optimal rules lead to falsification and allocative inefficiency. Optimizing the rule for allocative efficiency while accounting for falsification typically induces falsification, at a heavy cost to the agents. Moreover, falsification generates n

Carmen Oliver Huidobro, Mikolaj Czerkawski, Kevin Monsalvez-Pozo, Cesar Aybar

Pretraining generalist Earth observation foundation models requires data that combine several sensors, varied temporal sampling and, increasingly, text, yet open datasets rarely provide all three in one consistent, easily accessible form. We introduce MajorTOM:Elliot Pretrain, a collection of 279,166 tiles of 10 km drawn from the Major-TOM grid by clustering

Yuxuan Hou, Zhenghan Li

We study the role of design smoothness in the minimax estimation of conditional average treatment effects at a fixed covariate point. For design densities bounded above and away from zero, without further regularity, we derive upper and lower risk bounds with an arbitrarily small gap in the convergence-rate exponent. Under uniform design, we obtain matching

Leandro Carísio Fernandes, João Alberto de Oliveira Lima, Hudson de Martim, Eduardo Seiti de Oliveira

Long legal documents are often represented as single retrieval units, which can dilute relevant content and limit retrieval effectiveness. This study investigates structural segmentation as a document representation strategy to improve Brazilian legal document retrieval. Using the hierarchical structure of legal acts, we represent each document as a collecti

Jiajing Hu, Xu Han, Li-Ping Liu

Accurate uncertainty quantification is critical for many time series forecasting applications, and recent studies have increasingly adopted generative models to capture predictive uncertainty. Existing approaches often combine deterministic forecasting with generative modeling to improve predictive performance. However, their joint training is typically unst

Yuhui Hua, Shun-Jie Li, Chencheng Zhang

For a locally finitely generated Grothendieck category, we characterise the Gorenstein injective covering property by several equivalent conditions, including closure under coproducts. To complete this characterisation, we construct a countable commutative Noetherian Cohen--Macaulay local ring of dimension one for which this property fails, answering negativ

Alec Payne, Zihui Zhao

We extend Uhlenbeck's classical results on the generic properties of Laplacian eigenfunctions. In particular, we find a residual set of $C^k$ domains in $\mathbb{R}^n$, $k > n$, such that the nodal set of any Dirichlet eigenfunction intersects the boundary orthogonally along a $C^{k-1}$ embedded submanifold, whenever the intersection is non-empty. This i

Kefeng Liu, Pan Peng, Shengmao Zhu

We prove the Labastida--Mariño--Ooguri--Vafa conjecture for oriented links with arbitrary integer framings. Consequently, the framed reformulated invariants have finite integral LMOV expansions for every integer framing vector.

St John Grimbly, Nicolas Kuske, Evert A. Boonstra, Charel van Hoof

An embodied agent must prioritise among competing bodily needs. In active inference, it can respond to an urgent need by increasing the precision of information about that need or by increasing its preference for outcomes that relieve it. We compare these two interventions, separately and together, in a foraging gridworld with three lethal needs, and ask how

Yuyang Bai, Yu Wang, Yu Zhang

Searching text-rich knowledge graphs requires following relations while examining textual evidence. Predefined graph interfaces constrain what evidence agents can access, even when they write code to combine tool calls or process retrieved subgraphs. We propose direct knowledge graph interaction (DKGI), a harness in which language model agents create graph a

Yuhan Zhu, Xiangyu Zeng, Qingyi Si, Chenxu Yang

Visual agents must turn video observations into timely decisions to track progress and choose what to do next. Building on the emerging System One paradigm, we introduce VideoJev, a family of models that cast video understanding and interaction as structured decision tasks. We adapt pretrained multimodal large language models (MLLMs) by replacing autoregress

Ashutosh Hathidara

Controlling an acting LLM agent means deciding, at each step of a long-horizon task, whether to trust its next action, abstain, or hand off to a human. That decision needs a per-step confidence signal predicting whether the current rollout will eventually succeed. The agent's own self-reported verbalized confidence is the natural candidate, but existing

Jason Ken Adhinarta, Raina Wu, Ashutosh Sharma, Avinash Balakrishnan

Self-distillation is an emerging approach that trains a model to imitate its own predictions when conditioned on privileged information, such as correct answers or environmental feedback. Recent methods have reported better sample efficiency and generalization than reinforcement learning (RL), but it remains unclear whether these gains extend beyond in-distr

Yang Luo, Tiantian Ji, Zifeng Kang, Yucheng Wang

System One models such as Jev change how software makes decisions by mapping open-ended semantic contexts into structured judgments that directly drive program execution. However, this paradigm introduces a new security risk, which we term Jev Program Decision Hijacking (JPDH): attacker-controlled contexts can redirect program decisions through schema-valid

Huaiwen Pang, Jiale Han, Junying Chen, Wenjun Xiao

Legal question answering requires answers that are not only plausible, but also grounded in legally valid and citable evidence. Standard retrieval-augmented generation often relies on one-shot semantic matching, which is insufficient when legal evidence must be selected according to authority, applicability, and diversity across heterogeneous sources. We int

Kawai Chung, Chunkit Chan, Yuxuan Liu, Yauwai Yim

Dialogue and collaborative assistants repeatedly judge what a partner knows, believes, or intends. Typed decision models offer a potential alternative to generative large language models for bounded Theory of Mind (ToM) selection. We present Jev-ToM, comparing five typed decision models and seven generative baselines on 11,117 English text questions from nin

Chengtao Jian, Kai Yang

Direct Preference Optimization (DPO) has become a widely adopted approach for aligning vision-language models (VLMs), yet the security of multimodal preference data remains largely unexplored. Existing preference poisoning attacks either corrupt explicit supervision signals or require visible triggers, making them detectable under human inspection. We introd

Bruno M. Braga, Chris Gartland, Gilles Lancien, Eva Pernecká

A closed subspace $E$ of a Banach space $F$ is \emph{$L$-embedded in $F$} if there is a projection $P\colon F\to E$ such that $\|x\|=\|Px\|+\|x-Px\|$ for all $x\in F$. We show that if $X$ is a proper, geodesic metric space with locally finite $\H^1$-measure, then its Lipschitz-free space, denoted by $\F(X)$, is $L$-embedded in its bidual. We also extend an a

Yunuo Chen, Jingnan Gao, Bing He, Jinjing Zhao

Pretrained diffusion models provide powerful generative priors for extreme image compression, but their stochastic, iterative generation processes are not designed to decode compressed representations directly. We present PrismCodec, which recasts a pretrained Diffusion Transformer (DiT) as a bitstream-grounded one-step generative decoder. PrismCodec uses th

Haotian Liu, Yingjian Chen, Xiaohang Xu, Lang Huang

Large language models increasingly tackle tasks that require multi-step reasoning, where errors in intermediate claims can affect the final answer. Self-consistency (SC) improves reasoning by voting over sampled final answers, but its vote leaves intermediate disagreements unresolved. To address this problem, we introduce DDRG (Disagreement-Driven Verificati

Ziyang Ling

Large language models (LLMs) often diagnose full sight, static clinical cases well yet struggle in foresight, dynamic diagnosis. Prior work has improved this limitation by quantifying the information models acquire and by developing stronger follow-up strategies to improve diagnostic accuracy. In contrast, we introduce a distinctive hindsight to systematical

Zhouhan Zhong, Qiuhong Shen, Xingyi Yang, Jinyi Xu

3D part assembly aims to reconstruct a complete shape from a set of unassembled parts by predicting a rigid SE(3) transformation for each part. Because the placement of a part is rarely determined by its own geometry, the task demands reasoning over the entire part set, and existing methods, which rely on pairwise relations or iterative generation and verifi

Nutan Kumar Naik, Vijay Prasad Poudel, Aditya Kumar Saroj, Saurav Samantray

Benchmarks report what an Artificial Intelligence (AI) agent can do on a task suite, but not whether one released version has been engineered with the controls that make it reliable, safe and observable, or how strongly that is shown. This paper proposes AgentStripes, an evidence-graded maturity model for AI agents. It defines levels (0 to 7 stripes), proces

Rongtian Ye

Point-in-time (PIT) agents can recall post-cutoff facts even when their external evidence respects a historical cutoff. We examine whether this access predicts evidence gathering in 48 historical company-comparison tasks. Independent closed-book revenue probes identify the better-remembered company for each model on tasks with unequal recall, fixing the pred

Daniel Marín Pina, Michela Mapelli, Giuliano Iorio, Cecilia Sgalletta

Dormant black holes (dormant BHs, i.e., X-ray quiet binaries comprising a BH and a companion star) are probes of the effect of cluster dynamics and binary stellar evolution on compact-object binaries. A dozen dormant BH binaries are known, a number that will increase substantially with the upcoming Gaia Data Release 4 (DR4). Their origin is uncertain, and dy

Vladimir Bošković, Balázs Keszegh

We consider a natural extension of the classical saturation problem to oriented graphs, defined as the minimum number of edges in an oriented graph $H$ on $n$ vertices that avoids a fixed oriented graph $G$, but adding any missing edge to $H$ creates a copy of $G$. Our main result shows that saturation functions of oriented graphs attain at least four distin

Antonio De Rosa, Aria Halavati

We establish a Weyl law for the codimension-one anisotropic volume spectrum on compact Riemannian manifolds. We then use this law to obtain equidistribution in the full Grassmannian bundle while keeping the ambient metric fixed. More precisely, on every closed connected smooth Riemannian manifold of dimension three through six, we construct smooth, embedded

Fabian Gröger, Naël Ouerghemi, Shuo Wen, Maria Brbić

Multimodal LLMs are typically trained by first aligning a vision encoder with a language model via projector training on large image-caption datasets, followed by joint training of the full multimodal model. Despite its widespread use, the role of the projector training stage remains poorly understood. We find that (i) using only $10$ to $20\%$ of the data f

Peiwei Chang, Yongqiang Chen, Yuelan Cheng, Liting Gao

Physical properties provide intuitive controls for sounds made by interacting objects. The challenge is to make a generative model respond predictably to these controls, both individually and in combination. We present \method\ (PXS), trained on simulated contact sounds with 15 material, geometry and contact parameters. Two separately trained variants accept

Changyeon Kim, Kimin Lee, Yashraj Narang, Yijie Guo

Reinforcement learning (RL) post-training improves robot foundation models (RFMs) beyond their demonstrations, but real-world deployment demands more: learning new tasks efficiently while retaining previously acquired behaviors. Existing continual RL methods on top of VLAs mitigate forgetting but rely on on-policy updates with limited forward transfer, where

Chunyi Li, Zhiyu Liu, Alexander Perry, Xiaolei Zhao

We establish a Bogomolov--Gieseker-type inequality for tilt-semistable objects on smooth projective varieties of dimension $\geq 3$, in the spirit of the Bayer--Macrì--Toda conjecture. A key ingredient is an analysis of the action of the Frobenius morphism on tilt-semistable objects. As an application, we prove the Castelnuovo Bound Conjecture, up to an expl

Tobias Hürten, Johannes F. Loevenich, Florian Spelter, Erik Adler

This paper proposes an Autonomous Cyber Operations (ACO) gym, a unified framework integrating high-fidelity network emulation, realistic user behaviour modelling for benign network traffic and adversarial profiles, and the generation of cybersecurity scenarios through open toolkits. By supporting live and deferred feature extraction at the packet and flow le

Hilda Adwubi Osei

Fairness constraints in predict-then-optimize pipelines are necessarily enforced on predicted outcomes $\hat{y}$, since true outcomes $y$ are unobserved at decision time. We study whether a demographic parity constraint enforced this way continues to hold against $y$, and how the resulting violation scales with prediction error. Using a budget-constrained al

Per Lunnemann, Liselotte Jauffred

Every October, the Nobel Prizes in Physics, Chemistry, Physiology or Medicine and Economic Sciences are announced, with one to three prizes in each discipline. Of the 769 Laureates (since 1901) in these four disciplines only 4% are women (30 Laureates). In 2026, there are eight Laureates within these disciplines and they are all men. Based on historical data

Ziyi Li

We study aspects of the Euclidean partition function of the $\frac{SL(2,\mathbb R)\times U(1)}{U(1)}$ gauged WZW model. The target space geometry is either a smooth horizonless geometry with a linear-dilaton asymptotic region, or the Horne--Horowitz black string, depending on which circles we unwind as the time-direction. We first study the worldsheet spectr

Ilias Papagoudis, Dimitris Boskos

In this paper, we introduce a new class of data-driven ambiguity sets for distributionally robust optimization (DRO) that aim to enhance the local inference accuracy of an unknown distribution. Building on optimal transport-based discrepancy measures, we depart from the conventional paradigm of monolithic ambiguity balls, which rely on a single global transp

Diyi Liu, Shuchen Zhu, Wibe Albert de Jong, Di Fang

We construct Clifford+$T$ circuits implementing arbitrary $n$-qubit unitaries with worst-case $T$-count $\widetilde{O}(N^{8/7})$ at constant operator-norm error, where $N=2^n$, improving the previous exponent $5/4$. The construction uses $\widetilde{O}(N^{6/7})$ clean ancillas. The main idea is to reduce the data addressed by each query to the block-encoding

Irem Ecem Yelkanat, Faeze Ghorbanpour, Alexander Fraser

Annotator disagreement is increasingly recognized as a meaningful property of subjective tasks, rather than merely noise to be removed. Predicting which items are likely to be contested could help allocate annotation effort and route difficult cases to human review. We propose predicting disagreement from interpretable semantic properties of the text. An LLM

Ting Zhu, Huizhi Liang, Boguslaw Obara, Shufei Duan

Multimodal Emotion Recognition (MER) aims to understand human emotions by integrating information from multiple modalities such as speech and text. Psychological studies suggest that emotions exhibit latent hierarchical structured relationships in their semantic representations. However, most MER approaches rely on Euclidean representations, which capture fi

Amin Aboubrahim, Anastasiia Minenkova, Tomas Yonkers

The Klein-Gordon equation is a second order differential equation that appears in a variety of applications in mathematics, physics and engineering. This equation is often solved using numerical methods, since a general analytical solution is unattainable. However, one can find very useful analytical solutions for some special cases. In cosmology, the Klein-

José Afonso, João Madeira Pereira, Ronghao Ni, Limin Jia

JavaScript vulnerabilities in npm packages pose a significant risk to the Node.js ecosystem, and a growing body of work aims to detect, exploit, and mitigate them. Studying and evaluating this work requires up-to-date collections of real vulnerabilities, yet the most commonly used datasets stop recording vulnerabilities in 2022. We present X.js, a live bench

Zhichao Tang, Xiang Fan

For $f(X)=X^r(X^s+a)\in\mathbb F_{p^m}[X]$, with $m,r,s\geqslant1$ and $a\in\mathbb F_{p^m}^\times$, put $g=\gcd(s,p^m-1)$. We prove that, for each fixed $g$, only finitely many prime characteristics $p$ admit such permutation binomials, even as $m,r,s,a$ vary. If $6\nmid g$, every such $p$ divides Wendt's determinant $W_g=\operatorname{Res}_X(X^g-1,(X+1

Cong Hua, Qianqian Xu, Zitai Wang, Zhiyong Yang

Sampling-based test-time scaling enhances LLM reasoning by aggregating multiple reasoning paths for the same query. Existing methods use answer frequencies or path confidence to compute candidate scores, which do not explicitly characterize comparisons between competing answers. Inspired by the Condorcet principle in social choice theory, we introduce pairwi

Sayan Sengupta, Anish Sharma, George Biswas, Anindya Biswas

We investigate how unresolved quasistatic disorder in system bath coupling affects non-Markovian dynamics in spin environments. The coupling is fixed within each experimental shot but varies between shots, so repeated shot tomography reconstructs a disorder averaged dynamical map. For a single spin bath with pure dephasing Ising interaction, disorder can dri

Mark Whitmeyer

A seller without commitment to future offers interacts with a buyer over time. There are many goods for sale, with the buyer's value for a bundle the sum of each bundled good's value. Each date, the seller jointly announces a menu of subsets of the remaining inventory and current prices, and the buyer may purchase one of the offered subsets or wait.

Juan G. Calvo, César Herrera

We introduce and analyze a two-level overlapping additive Schwarz preconditioner for discontinuous Galerkin discretizations of virtual element spaces on general polygonal meshes in two dimensions, for a second-order elliptic problem whose coefficient is constant on each subdomain and may have large jumps across the interface. The discretization is a weighted

Yubo Li, Ramayya Krishnan, Rema Padman

System One models choose an agent's next action by scoring a supplied set of options in a single pass. This makes them fast and cheap, but when they choose wrongly there is nothing to inspect: did the model misread what the agent has seen, or read it correctly and still act wrongly? We answer behaviorally. On a frozen benchmark of agent decision states f

Yuwen Cao, Hui Wang, Tomoaki Ohtsuki, Shahid Mumtaz

Rate splitting multiple access (RSMA) has emerged as a promising non-orthogonal transmission mechanism to improve spectral- and energy efficiency, and user fairness in multi-antenna broadband channels. To further enhance its performance, in this paper we investigate a novel cooperative RSMA (C-RSMA) transmission framework assisted by reconfigurable intellige

Bo Xue, Ji Cheng, Yuanyu Wan, Shuang Qiu

Best-arm identification (BAI) aims to identify the optimal arm using as few samples as possible. While most existing work focuses on a single objective, many real-world applications involve multiple criteria, which motivates the study of BAI in multi-objective bandits. Intuitively, multi-objective identification seems harder, since the learner must identify

Suchira Sarkar, Kanak Saha, Anshuman Borgohain

The massive disk galaxies (stellar mass >= 10^11 M_solar ) in the local universe (z < 0.1) encode challenging aspects of galaxy formation physics. We have studied one class of such galaxies - the giant low surface brightness galaxies (gLSBs) that show a hybrid morphology - a central Sersic, an inner high surface brightness, exponential stellar disk surrounde

Ananya Kotla, Jorge T. Gomez

Post-training INT8 quantization is typically validated by checking accuracy retention, which requires ground-truth labels often unavailable in low-resource clinical settings. We show a label-free alternative: comparing a model&#39;s own output before and after quantization. We compute cosine similarity between the FP32 and INT8 models&#39; output probability

Jia Wei, Muhuo Liu, Zhifu You, Hong-Jian Lai

Let $τ(G)$ denote the maximum number of pairwise edge-disjoint spanning trees contained in a graph $G$. Zhang and Fan [J. Graph Theory, 2026] posed the problem of determining the corresponding tight spectral-radius condition for every $s$-edge-connected graph with \[ k+1\leq s\leq2k-1, \] together with a characterization of the associated extremal graphs. In