Probability
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Thursday 3 September
This paper is the third and final article in a series of papers constructing the canonical conformally invariant metric on the set of loops of the conformal loop ensemble (CLE) with critical parameter $\kappa=4$. The previous two articles construct, as a subsequential limit of the renormalized graph metric on the loops of CLE$_\kappa$ as $\kappa \downarrow 4$, a conformally invariant, local metric on the loops of a CLE$_4$ whose metric ball growth from the domain boundary coincides with the uniform exploration of Werner and Wu. In this paper, we establish that this metric is uniquely characterized by its properties, as are its geodesics, and that it is a measurable function of the CLE$_4$. In particular, we show that the renormalized CLE$_\kappa$ graph metric converges as $\kappa \downarrow 4$ without passing to a subsequence. A key step in the proof is to show that the metric is determined by the geodesics from each loop to the domain boundary, which are in turn determined by the uniform exploration; this representation will have important applications in future work.
Probability
The conformally invariant metric on CLE$_4$ II: existence of geodesics
We continue our study of the conformal loop ensemble (CLE) with parameter $\kappa=4$, the critical threshold at or below which the loops are simple and disjoint, touching neither each other nor the domain boundary. This paper is the second in a series of three establishing that the loops of a CLE$_4$ uniquely determine a conformally invariant, local, and geodesic metric such that the metric ball growth from the domain boundary coincides with the uniform exploration of Werner and Wu. In this second paper, we prove the existence of geodesics, showing that any geodesic between two loops is supported on the CLE$_4$ loops (off a set of Hausdorff dimension zero) and does not intersect the domain boundary. Along the way, we establish sharp quantitative estimates for the CLE$_4$ metric geometry, including exponential tail bounds for rectangle distances and multi-scale four-arm SLE$_4$ non-intersection bounds for metric balls.
Read the full abstract
Probability
The conformally invariant metric on CLE$_4$ I: subsequential limits of the non-simple CLE graph metric
We consider the conformal loop ensemble (CLE) with the parameter $\kappa=4$, the critical value at or below which the loops are simple and do not intersect each other or the domain boundary. We show that the loops of a CLE$_4$ uniquely determine a conformally invariant, local, and geodesic metric so that the metric ball growth from the domain boundary coincides with the uniform exploration of Werner and Wu. This metric was previously constructed in unpublished work of Sheffield, Watson, and Wu. Our approach differs in that we show that the metric arises as the renormalized limit of the graph metric on CLE$_\kappa$ loops as $\kappa \downarrow 4$. In this first paper in a series of three, we prove that the subsequential limits exist and define a non-trivial conformally invariant metric on CLE$_4$ which is local and such that the metric ball growth from the boundary is given by the uniform exploration of Werner and Wu. In subsequent work, we will show that the subsequential limit exists as a true limit.
Read the full abstract
High Energy Physics (Theory)
Boundary conditions for axionic wormholes, imaginary distance bound and KSW allowability
We study four-dimensional axion wormholes in the Lorentzian mini-superspace path integral using dual scalar and three-form flux formulations. We analyze this duality for generic metric boundary conditions and real lapse integration contour, and show that it holds on any background, including complex ones. With the Dirichlet condition on the metric, the fixed-flux path integral is evaluated exactly. In the Euclidean regime, saddle geometries organize into same-side and cross-throat segments of the Giddings-Strominger (GS) wormhole, depending on whether the two boundaries lie on the same or opposite sides of the throat. A Picard-Lefschetz analysis in the covering plane of lapse reveals that the cross-throat saddle has vanishing intersection number, and therefore does not contribute. In the asymptotically flat limit, it corresponds to the imaginary wormhole that saturates the imaginary distance bound (IDB). The same conclusion follows independently on the scalar side of the duality, and is confirmed by the exact amplitude. Replacing the Dirichlet condition with a one-parameter Neumann condition, the cross-throat saddle that leads to the imaginary wormholes remains irrelevant for the physical lift of the contour. This purely imaginary parameter characterizing the Neumann condition interpolates continuously between the half- and complete-wormhole geometries. Convergence of the sum over fixed-charge sectors imposes a corresponding interpolating bound not only on the imaginary part of the boundary axion but also on the parameter characterizing the Neumann boundary condition. An independent analysis based on the KSW criterion reproduces exactly the same bound, emphasizing its compatibility with the IDB.
Read the full abstract
Statistics Applications
Natural Disasters and the Nonprofit Sector
When natural disasters strike, individuals, communities, and even entire countries can suffer. Researchers have studied the impacts of disasters on various factors of interest, from mental health, to poverty, to economic activity. However, the impact of disasters on the nonprofit sector is understudied despite the nonprofit sector's perhaps surprising role in local or national economies as well as its role in disaster response and recovery. Thus, we study the effect of natural disaster damage on different county-level nonprofit outcomes using a panel dataset spanning 1991 to 2021 and causal inference methods tailored to panel data. Contrary to prior work, which found small but positive associations between disaster damage and nonprofit revenue or assets, we find no evidence of a causal effect.
Read the full abstract
Artificial Intelligence
The Natural Language Interaction Protocol and Standard for AI Agents
AI agents are increasingly being developed and deployed across organizations using heterogeneous agent-development frameworks, AI models, tool interfaces, protocols, and execution environments. To realize their potential social and business impact, these agents must be able to interoperate through a common communication protocol. The Natural Language Interaction Protocol (NLIP), developed by researchers and practitioners across companies and universities and standardized by Ecma International, addresses this need by defining a standards-based application-layer protocol for AI-agent interaction. NLIP provides a lightweight semantic message envelope that can be carried over existing transports such as HTTP/HTTPS, WebSocket, and AMQP, while allowing NLIP-aware agents and gateways to adapt between clients, agents, local context stores, ontologies, tools, enterprise services, and heterogeneous underlying protocols. This paper presents the motivation and design rationale of NLIP, its message model and transport bindings, security-by-design considerations, reference implementation, representative applications, adoption signals, and relationship to emerging agent protocols such as MCP and A2A.
Read the full abstract
Machine Learning
Prospective Coding Improves Learning in Deep Continuous-Time Recurrent Networks
Temporal integration gives continuous-time recurrent networks memory, but in deep stacks it also delays bottom-up signals and attenuates top-down errors. We compensate for this delay by making each layer's bottom-up input prospective, and we develop Recursive Quadrature Filters (RQFs), biologically motivated temporal band-pass filters that are a special case of diagonal state-space models (SSMs). Starting from an energy model, we derive the RQF dynamics and show that each RQF is a band-pass filter whose learnable parameters control its tuning frequency and bandwidth. We then make each layer's bottom-up input prospective using a parameter-free two-tap update that leaves the recurrent transition and parallel scan unchanged. We extend this correction to general diagonal SSMs and show that it mitigates depth-dependent gradient attenuation when temporal gradients are truncated, i.e., spatial-only backpropagation. We evaluate the performance of RQFs, S5, and ORGaNICs (a nonlinear gated RNN with built-in attentional modulation and recurrent normalization) trained using full backpropagation through time (BPTT) and spatial-only backpropagation. Under full BPTT, the prospective variants of these models match or outperform their non-prospective controls in every configuration. A non-residual width-32 six-layer RQF achieves 96.09% accuracy on raw-audio Speech Commands with 31.9k parameters; a width-64 six-layer RQF achieves 83.36% on the 16,384-step Path-X task. We also introduce selective RQFs (sRQFs), which make each filter's input gain and, optionally, its bandwidth and tuning frequency input-dependent while retaining the parallel scan. With the input gain alone, the same six-layer model achieves 86.31% on Path-X; with all three input-dependent, a four-layer prospective sRQF solves Selective Copying (99.83%).
Read the full abstract
Cosmology and Nongalactic Astrophysics
The moving lens effect: analytical modelling and foreground suppression
The moving lens (ML) effect is a secondary anisotropy of the cosmic microwave background (CMB) generated by the transverse motion of gravitational potentials, providing a direct probe of the large-scale cosmic velocity field. Its detection relies on cross-correlating a CMB map with the transverse galaxy momentum field, constructed from the galaxy overdensity and a velocity field reconstructed from it. Restricting the reconstruction to large-scale modes strongly suppresses contamination from small-scale astrophysical foregrounds while preserving the ML signal. In this work, we develop a theoretical framework for the ML estimator and its foreground contamination. We show that the signal and foregrounds have a distinct dependence on the direction of the long-wavelength mode represented by the reconstructed galaxy velocity. In the squeezed limit enforced by the velocity reconstruction filter, the bispectrum sourcing the foreground correlation becomes independent of this direction, causing its leading contribution to vanish after angular averaging. In turn, the ML signal survives by matching this directional dependence, with its amplitude reduced only by the filtered velocity variance. Using the halo model, we derive the leading corrections beyond the squeezed limit and show that the residual contamination remains parametrically suppressed. Finally, we model the cross-correlation exactly in the curved sky, and show that it is sourced solely by the longitudinal component of the galaxy momentum field, in the form of a spin-1 $E$-mode, with all other contributions either strongly suppressed on small scales or exactly zero.
Read the full abstract
Quantum Physics
Characterizing Large Scale Quantum Systems with Error Per Circuit Layer
Quantum benchmarks provide compact measures of performance that are important for evaluating and comparing quantum systems. Circuit-level benchmarks are particularly valuable because they capture the accumulated effects of noise across interacting operations, but existing approaches may require structured gate sets and costly compilation, classical simulation of reference outputs, or subsystem decompositions that do not capture full-register behavior. We introduce Error Per Circuit Layer (EPCL), an overlap-based circuit-level benchmark that estimates an effective layer polarization by applying identical random circuits to two disjoint quantum registers and measuring the overlap between their output states as a function of circuit depth. EPCL avoids classical simulation of ideal output distributions and recovery to a known reference state, and is compatible with arbitrary gate sets, including non-Clifford gates. We derive the expected overlap decay under an ensemble-averaged depolarizing model and identify the assumptions under which the fitted decay parameter represents an effective layer polarization. Numerical simulations show that EPCL recovers the predicted polarization under weak local stochastic noise and remains well described by a single-exponential decay at stronger stochastic noise levels. The simulations further show that coherent errors associated with fixed entangling layers may require Pauli twirling or randomized compiling to produce the expected decay, while inter-register correlations contribute an additional covariance term to the measured overlap. Finally, experiments on IBM quantum hardware demonstrate clear EPCL decay in 8- and 16-qubit implementations. These results support EPCL as a method for measuring aggregate register performance without requiring classical simulation of ideal circuit outputs or restriction to structured gate sets.
Read the full abstract
Computer Vision
Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding
Streaming video understanding requires multimodal large language models (MLLMs) to process continuous visual inputs and respond to user queries under strict causality and bounded memory. Existing approaches typically compress historical observations into an external memory bank and retrieve query-relevant evidence as additional visual context. Though effective, this store-and-retrieve paradigm keeps historical evidence as external visual context, preventing it from being internalized into a compact, evolving latent memory that can continuously guide streaming reasoning. To bridge this gap, we introduce LatentStream, a progressive latent working memory framework that shifts streaming memory from store-and-retrieve to retrieve-and-internalize. Specifically, LatentStream comprises three coordinated components. First, Query-agnostic Hierarchical Streaming Memory organizes visual history into short-, mid-, and long-term levels under a fixed memory budget through Jenks-guided adaptive consolidation. Once a query arrives, Hierarchical Latent Memory Evolution equips groups of latent memory tokens with progressively expanding memory receptive fields, enabling them to iteratively retrieve historical evidence from their corresponding scopes and internalize it into a compact, fixed-length latent memory. Finally, Progressive Confidence-guided Latent Memory Optimization constructs a hierarchical progression reward from group-wise predictive entropy and jointly refines the latent memory tokens and retrieved evidence, encouraging increasingly confident streaming reasoning. Extensive experiments demonstrate that LatentStream achieves new state-of-the-art results on existing online and offline video benchmarks.
Read the full abstract
Astrophysics of Galaxies
Fundamental differences in the X-ray accretion properties of low and high-excitation radio galaxies
We characterise the accretion rate properties of radio-detected AGN by combining deep radio and X-ray observations of the Boötes field. We used deep international LOFAR telescope observations to identify 2840 radio-AGN across $0.3 < z \leq 2$, divided into samples that are complete in radio luminosity. We further split this sample into four different classes: radio-quiet AGN (RQ-AGN), high-excitation radio galaxies (HERGs), and low-excitation radio galaxies (LERGs) hosted by star-forming (SF-LERGs) and quiescent galaxies (Q-LERGs). Through performing X-ray stacking, we determined the average X-ray luminosities, $L_{\rm{X, 2-10\,keV}}$ and average specific X-ray luminosities, $\lambda_{sL_{X}}$ (X-ray luminosity scaled by the stellar mass; a proxy for the accretion rate). We studied how these X-ray properties depend on radio luminosity, stellar mass, and redshift for each of the four AGN classes. We found that the LERGs, regardless of their star-formation activity, show significantly lower $L_{\rm{X, 2-10\,keV}}$ and $\lambda_{sL_{X}}$ than both HERGs and RQ-AGN across all redshifts. The average X-ray luminosities for the HERGs, RQ-AGN, and SF-LERGs typically increase with redshift, which may be associated with the increased cold gas fractions at earlier times, resulting in more enhanced black hole accretion. The average X-ray luminosities show weak-to-no correlation with radio luminosity at a given redshift, suggesting that the physical processes producing the X-ray and radio emission may not be coupled on spatial and/or temporal scales.
Read the full abstract
Lattice Field Theory
Lattice QCD calculation of the pion-nucleon coupling $\bar{g}_0$ induced by the QCD $\Theta$-term
We present lattice QCD results for the CP violating pion-nucleon coupling $\bar{g}_0$ induced by the QCD $\overline\Theta$ term from the analysis of three 2+1+1-flavor ensembles generated with highly improved staggered quarks (HISQ) by the MILC collaboration. These ensembles are at lattice spacing $a\approx 0.09~\text{fm}$ and pion masses of 313, 226 and 138 MeV, respectively. The coupling $\bar{g}_0$ is extracted in two ways. First, from the matrix element of the correlation between the pesudoscalar current and the topological charge evaluated between the nucleon ground state. The data for correlation functions with both the pseudoscalar and axial vector exhibit large contamination from the $N\pi$ excited state. We show that these can be controlled at the leading order using chiral perturbation theory ($\chi$PT) and the axial Ward identity (AWI, also called the partially conserved axial current (PCAC) relation). The result after removing the $N\pi$ contamination and extrapolating to the physical pion mass is, however, noisy: $\bar{g}_0/(2F_\pi)=-7(63)\times 10^{-3} \,{\overline{\Theta}}$. The more precise result $\bar{g}_0/(2F_\pi)=17.4(1.9)\times 10^{-3} \,{\overline{\Theta}}$ is obtained using low energy effective field theory methods or equivalently the AWI. Since contamination from the $N\pi$ excited states arises in the calculation of many nucleon matrix elements, we give an extended discussion on them and the use of the AWI for controlling them in the calculation of $\bar g_0$.
Read the full abstract
Artificial Intelligence
Environment Evolution for Terminal Agents
Scaling interactive and verifiable environments is critical for training terminal agents. As frontier models become more capable, environments synthesized from scratch become less challenging and thus provide limited learning signals. Recent co-evolution methods iteratively synthesize environments near the model's learnable frontier based on weaknesses exposed during rollouts. However, their dependence on on-policy rollouts limits generalization and the continuous provision of learning signals as the model becomes stronger. In this paper, we propose environment evolution, which incrementally increases environment difficulty off-policy and schedules the evolved environments generation by generation during training to provide continuous learning signals. We derive three evolution directions that influence environment difficulty from the multi-turn learning objective and then implement evolution along these directions through a loop-engineered multi-agent harness. Quantitative rollout experiments with Hy4 preview, Claude Opus 5, and GPT-5.6 Sol show that environment evolution consistently produces more difficult environments. We validate its effectiveness on Qwen3.6-27B and Qwen3.6-35B-A3B through simple long-horizon RL training, improving their performance by 14.4 and 18.0 percentage points on Terminal-Bench 2.1, respectively.
Read the full abstract
Artificial Intelligence
Epistemic Warrant for LLM Recommendations: Characterizing the Basis for Reliance When Ground Truth Is Unavailable
Large language models are increasingly used to support organizational decisions. Despite their rapid and widespread adoption, users often lack a principled basis for assessing whether to rely on a specific recommendation. We introduce epistemic warrant, a decision-level construct that characterizes the scope over which a recommendation is supported. We operationalize this construct through a reliance certificate that uses behavioral evidence from four tiers of tests to assess a recommendation's underlying warrant and assign it to a warrant category. We assess this operationalization using complementary forms of construct-validity evidence, including transformation validity, known-groups tests, independent human consensus, discriminant analyses, and robustness across models and alternative implementations. The results show that the certificate behaves in theoretically expected ways: its transformations instantiate their intended relationships, certificate assignments increase across groups prespecified to represent increasing levels of warrant, and stronger assigned warrant is associated with greater independent human consensus. Certificate assignments also capture information beyond verbalized confidence and are not readily explained by decision difficulty. The main conclusions remain robust across models and operationalizations. These findings support the reliance certificate as an operationalization of epistemic warrant for assessing the scope of support for a particular LLM recommendation when ground truth is unavailable.
Read the full abstract
Statistical Computation
Fast Computation of Nested Cross-Validation for Penalized Regression
Cross-validation is a resampling procedure that provides a point estimate of generalization error for any predictive model. Cross-validation is widely used for model selection and evaluation. Uncertainty in the cross-validation estimate is challenging to quantify, and estimation of its variance is known to require multiple runs of the entire resampling procedure. Nested cross-validation computes a prediction interval for the generalization error for a given model and training dataset by resampling the entire cross-validation procedure but incurs extraordinary computational cost. We provide an efficient method for computing the nested cross-validation prediction interval using only a single model fit for some penalized regression models including ridge regression, spline smoothing, and some functional regression models. We characterize when our proposed method should be expected to out-perform resampling-based nested cross-validation in various scaling regimes as well as in finite samples. Experiments for functional principle components regression demonstrate non-trivial cases in which our proposed method improves run times substantially.
Read the full abstract
Computers and Society
Decreasing Digital Distraction in College Students: Associated Online Learning Strategies Identified by Unsupervised Data Mining Approaches
The proliferation of digital tools in education offers numerous benefits but also introduces significant challenges, notably digital distractions that hinder academic performance, especially in online learning contexts. This study employed unsupervised data mining techniques, specifically association rule mining and clustering analysis, to identify effective learning strategies associated with lower levels of digital distractions among college students. Data from 530 participants revealed that self-regulated learning strategies (i.e., goal setting, environment structuring, and time management) co-occurred most consistently with lower digital distractions. Additionally, learner-instructor and learner-content engagement strategies, as well as technical competencies, also tended to appear in the same profiles as lower distraction. Interestingly, reliance on peer help-seeking and learner-learner engagement strategies appeared less often in those lower distraction profiles. These findings offer actionable implications for educators to design targeted interventions that foster focused and productive online learning environments.
Read the full abstract
Statistics Theory
Bernstein–von Mises theorems for Bayesian probabilistic numerics
We study probabilistic numerical methods for solving nonlinear PDEs from a Bayesian nonparametric perspective. Given noisy evaluations at random collocation points, we place a truncated Gaussian series prior on the unknown solution and establish contraction at the minimax nonparametric rate, up to a logarithmic factor. Our main results give Gaussian approximations of the posterior in positive-order Sobolev spaces and, under suitable conditions, in the uniform topology. This contrasts with classical ill-posed inverse problems, where Bernstein–von Mises theorems typically require substantially weaker topologies. Here, the observation operator is differential rather than smoothing, and inversion of its linearisation gains regularity, making these strong-topology results possible. The posterior may be centred at either the posterior mean or the posterior mode. We further prove that the Gaussian Laplace approximation is asymptotically equivalent to the true posterior at a $\sqrt{N}$-scale.
Read the full abstract
Software Engineering
ATIBA: Grounded Integrity and Quality Checking for Research Papers
Checking a manuscript's reference integrity, its compliance with a target venue's specific submission rules, and its adherence to community reporting standards is manual, repetitive, and different for every venue so in practice it is done inconsistently or skipped. We present ATIBA, a tool that runs five grounded integrity and quality checks on a manuscript: a reference-integrity check that verifies each citation against bibliographic sources and flags retracted or unfindable references; a venue/track compliance check that derives submission criteria directly from a venue's own call-for-papers page and evaluates the manuscript against them, each verdict anchored to a verbatim quote from that page; an empirical-standards compliance check against the ACM SIGSOFT Empirical Standards, with a hallucination defence that discards any evidence quote it cannot locate verbatim in the manuscript; a multi-mode AI review (venue-specific, formal, and page-anchored annotation) powered by GPT-5.4 through Azure OpenAI; and a citation-suggestion feature that proposes candidate references for a manuscript and verifies each against bibliographic sources before it is shown to the user. All five checks are designed around the same principle: an LLM is only trusted to judge, never to invent the evidence it judges against. We evaluated ATIBA through a moderated user study with 13 non-author participants. Agreement across the six survey items ranged from 69% to 92%, with a mean of 85%, providing initial evidence of positive perceived usefulness across the evaluated workflows. These findings establish perceived usefulness; objective accuracy remains to be measured.
Read the full abstract
Information Theory
Synchronization Strings over the Optimal Alphabet
Synchronization strings provide deterministic position labels for recovering coordinates after insertions and deletions. Haeupler and Shahrasbi introduced these objects, and subsequent work proved that four symbols suffice for some fixed parameter epsilon < 1, whereas two symbols cannot support arbitrarily long synchronization strings. We resolve the remaining ternary case: every length admits a ternary 2001/2002-synchronization string. Thus three is the exact minimum constant alphabet size. A computer-assisted refinement based on a larger 54-uniform family yields ternary epsilon-synchronization strings for every epsilon > 215/216. The previous four-symbol construction uses a ternary square-free backbone to exclude short repetitions and a fourth symbol to carry long-range synchronization marks. Our main technical contribution is a local-entropy transfer theorem: every square-free block-local source with a positive interval conditional min-entropy rate supports synchronization strings with a fixed gap. We instantiate this theorem using occurrence-wise branching in a Brinkhuis family. Every outcome remains ternary and square-free, while every long interval retains linear conditional min-entropy after all choices outside it are exposed. A deletion-ball estimate converts this entropy into an exponentially small probability of a near-complete common subsequence between adjacent intervals, and an asymmetric Lovasz Local Lemma enforces all interval constraints simultaneously. The same framework also yields exponentially many valid words, synchronization circles, and synchronization within a class of extremal square-free words. Adding constraints on distant intervals gives a Las Vegas construction in expected O(n^2 log^3(n+2)) time.
Read the full abstract
Quantum Physics
Quantum thermalization achieves optimal approximate quantum error correction
Quantum thermalization explains how an isolated many-body system naturally evolves towards a thermal state, rendering information about the initial conditions inaccessible to local measurements. This is precisely the mechanism utilized in quantum error correction, where information is protected by design through a nonlocal encoding. In this work, we leverage this connection to port the rigorous framework of (approximate) quantum error correction to the study of quantum thermalization. Treating typical late-time states as codewords, we characterize the error-correcting properties of generic thermalizing dynamics. We numerically uncover a universal relationship between the encoding rate, distance, and thermal entropy density of the emergent code. At infinite temperature, this universal curve saturates the quantum Singleton bound, achieving the same optimal limit as Haar-random codes. At finite temperature, we introduce a code family based on the Scrooge ensemble, the natural thermal analogue of the Haar ensemble, and prove it saturates the entropic quantum Singleton bound, establishing this family as optimal within entropic constraints. Our extracted universal curve independently saturates this same bound, revealing that finite-temperature thermalization is itself optimal. Finally, we show how conserved quantities limit the error-correcting behavior of thermalization: codewords with differing energies, or other conserved charges, leak only classical information, and correctability persists until the difference reaches the scale of thermal fluctuations. Our results reveal a universal optimal coding structure in thermalizing dynamics, while introducing new optimal codes that achieve fundamental limits of approximate quantum error correction.
Read the full abstract
Computer Vision
BooM-VVT: Boosting Mask-Free Video Virtual Try-On with Image-Level Pseudo Data
Video virtual try-on (VVT) aims to generate realistic videos of a person wearing a target garment. Recent methods leverage a keyframe-driven video generation paradigm to improve in-the-wild performance, yet they still rely on masks to localize try-on regions, making them vulnerable to large motions and severe occlusions. Although mask-free image-based try-on methods have shown promising results by leveraging large-scale pseudo data, extending this paradigm to videos remains difficult, as constructing video-level pseudo data is prohibitively expensive. Furthermore, coarse keyframe sampling and the scarcity of multi-view try-on data limit existing keyframe-driven methods in maintaining garment consistency and handling diverse try-on tasks. To address these challenges, we propose BooM-VVT, a mask-free VVT framework built upon the keyframe-driven paradigm. To achieve mask-free VVT, we introduce a multi-stage training strategy that leverages image-level pseudo data for mask-free localization learning, substantially reducing the need for costly video-level pseudo data. To improve garment consistency, we propose Garment-Sensitive Keyframe Sampling, which selects keyframes based on garment-relevant body regions to better capture garment appearance. We further introduce Frame-Shared 3D-RoPE to establish spatiotemporal correspondences between keyframes and target video frames for accurate garment-detail transfer. Finally, we construct OmniView, a large-scale multi-view try-on dataset to support reliable try-on video generation under complex camera viewpoints and diverse try-on tasks. Extensive experiments demonstrate that BooM-VVT achieves superior temporal consistency and garment fidelity over existing methods. Project page: https://boomvvt.github.io/boomvvt.
Read the full abstract
Quantum Physics
Non-local Magic: closed-form solution and equivalence with magic of purification
Non-local magic quantifies the non-stabilizerness beyond the local choice of basis. Its evaluation is a hard problem requiring optimization over local unitaries. Here, we show that for the log-stabilizer fidelity this optimization admits an exact closed-form analytic solution depending only on the Schmidt spectrum. We then address the open problem of non-stabilizerness of mixed states by introducing the magic of purification and show that it naturally induces a resource theory whose free states are the states with zero stabilizer entropy. For the log-stabilizer fidelity, the magic of purification admits a distance-based formulation in terms of the Uhlmann fidelity. We prove that non-local magic coincides with the minimum magic of purification along the unitary orbit of the reduced density operator. This framework introduces non-local magic as the relevant resource for non-trivial steering and the characterization of entangling and non-stabilizer resources in measurements. These results provide both an efficient analytical characterization and a mixed-state resource-theoretic interpretation of non-local magic.
Read the full abstract
High Energy Astrophysical Phenomena
A targeted search for fast radio bursts from magnetars in elliptical galaxies
Motivated by predictions that globular clusters associated with nearby giant elliptical galaxies are prime environments for detecting fast radio bursts (FRBs) from magnetars, we report on a targeted high-sensitivity search across five local massive elliptical galaxies. We utilized the GREENBURST single-pulse search pipeline and the VEGAS backend on the Green Bank Telescope (GBT) to conduct an L-band (1.4 GHz) survey for FRBs in each target environment. We discovered FRB 20240817A in the direction of M49 (NGC 4472) which has a pulse width of 15 ms, a signal-to-noise ratio of 13, and a dispersion measure (DM) of $983~\rm{cm}^{-3}$ pc. Although a Bayesian spatial association framework initially flags M49 as the host, carefully accounting for the foreground electron column contributions from the Milky Way and both the Virgo intracluster medium and M49's hot gaseous halo ($\sim300-400~\rm{cm}^{-3}$ pc combined) strongly disfavours a physical connection. We conclude that FRB 20240817A is a background event. We also identified a repeating cluster of three low-significance pulses from the direction of M49 at a physically plausible DM of $150-160~\rm{cm}^{-3}$ pc. Accounting for the fraction of each target's extended globular cluster distribution observed, the absence of unambiguous, host-localized bursts allows us to place a joint upper limit on the intrinsic burst rate of a single active source within these systems of $R_0 < 0.028~\rm{hr}^{-1}$ (and an integrated rate limit of $<1.0$~hr$^{-1}$ for M49), lowering previous upper limits and suggesting that magnetar formation within dynamical clusters is less efficient than previously assumed.
Read the full abstract
Lattice Field Theory
Non-Abelian string melting and thermalization in an open lattice gauge theory
Open-system lattice gauge theory (LGT) has so far been developed predominantly in Abelian settings, leaving open how genuinely non-Abelian gauge structure reshapes dissipative real-time dynamics. Here, we study a $1+1$D SU(2) Yang–Mills LGT with dynamical matter coupled to a thermal scalar environment through a gauge-preserving Lindblad evolution, which we solve using tensor networks. Starting from a quark–antiquark pair connected by a chromoelectric flux string, we find that the thermal medium melts the string by delocalizing the color charges and screening the flux; on resonance, this dissipative melting competes with and delays coherent string breaking. The thermalization time is non-monotonic in the environment coupling, decreasing through environment-assisted transport at weak dissipation before increasing in a quantum-Zeno regime. In the strong-dephasing limit, a Schrieffer–Wolff expansion maps the dynamics to a classical exclusion process and yields the Liouvillian thermalization time analytically. Beyond these generic open-system effects, the non-Abelian matter structure produces a systematic mesonic bias in the steady state, while the thermalization time decreases with temperature, in contrast to the Abelian Schwinger model trend and in qualitative agreement with pNRQCD studies of the quark–gluon plasma. These results establish a gauge-preserving framework for thermalization and string dynamics in open non-Abelian lattice gauge theories.
Read the full abstract
Optimization and Control
A Stochastic Riemannian Alternating Descent Ascent Method for Nonsmooth Composite Expectation Optimization on Riemannian Manifolds
In this paper, we consider a class of Riemannian nonsmooth composite expectation optimization problems, which arises in various machine learning, signal processing, and statistics applications. Noting that these problems admit structured minimax reformulations, we propose an efficient algorithm, named stochastic Riemannian alternating descent ascent method with recursive momentum (StoRADA-RM), to tackle them. StoRADA-RM performs one or multiple Riemannian stochastic gradient descent steps and then a proximal gradient ascent step at each iteration. To compute the Riemannian stochastic gradient, we propose a vector transport-free recursive momentum estimator that requires only $O(1)$ stochastic gradient evaluations per iteration. We prove that StoRADA-RM returns an $\epsilon$-Riemannian-stochastic-stationary point of a given problem in the said class in $O(\epsilon^{-3})$ iterations while making $O(\epsilon^{-3})$ calls to a stochastic first-order oracle (SFO). Both the iteration complexity and SFO complexity bounds are the best known in the literature for the said class of problems. The latter even matches the optimal lower bound for smooth nonconvex optimization with stochastic first-order algorithms. We then present numerical results on sparse principal component analysis and coordinate-independent sparse estimation to demonstrate the superior performance of our proposed method.
Read the full abstract
Statistics Theory
One Probability, Two Roles: The Separation of Coherence and Frequency in Adaptive Regimes
Probability plays two distinct roles in modern data-analytic practice: (i) as an internally coherent, filtration-relative language for sequential forecasting and, together with a stated loss or utility, decision making; and (ii) as a foundation for empirical claims such as stabilization, calibration, and repeated-sampling validity. The first role is relative to an assessment law $\mathbb Q$, whereas the second is evaluated under a governing law $\mathbb P$ and a declared repetition regime. In classical settings the roles are often aligned by conditional correctness under $\mathbb P$ together with i.i.d.\ or other stable-law assumptions under a fixed design, or by exchangeability. In adaptive and AI-mediated settings they can diverge because forecasts operate within feedback loops while reliability is evaluated under a potentially different law and regime. We synthesize relevant results from prequential forecasting, calibration, martingale and game-theoretic validity, adaptive data analysis, conformal prediction, and long-context AI reliability into a formal account of role separation. The resulting Role Separation Principle distinguishes internal coherence from conditional-mean correctness under the evaluation law and from target-specific reference and stability conditions. Two separation results show that a proper coarsening of the limiting-frequency distribution need not identify the governing law, and that even conditionally correct forecasts under that law need not accompany stabilization. We then develop a unified stabilization framework spanning filtration-based, shift-based, policy–environment, and design-based regimes and operationalize it through a regime-conditional audit checklist. Applied illustrations clarify calibration claims, synthetic-data validity, and structural mismatches relevant to some high-confidence unsupported AI outputs.
Read the full abstract
Lattice Field Theory
Real-Time String Dynamics in $3+1$D Lattice Quantum Electrodynamics
Understanding real-time string dynamics in three spatial dimensions is essential for connecting quantum simulations of lattice gauge theories (LGTs) to the physical dimensionality of QED and QCD, where transverse fluctuations and competing local processes proliferate. We present the first real-time simulations of string breaking in $3+1$D lattice quantum electrodynamics. Using tree tensor networks, we simulate the quench dynamics of electric flux strings in a $3\!+\!1$D U(1) LGT with dynamical matter. At strong coupling, the string breaks resonantly at a sharp resonance condition of mass and gauge couplings, converting electric energy into matter–antimatter pairs that screen the static charges. Off resonance, we classify all competing channels—pair production, string deformations and extensions, and flux loops—whose multiplicity, extensive for pair production and flux loops, depletes the string sector even far from resonance. A channel-resolved perturbation theory quantitatively reproduces these dynamics and their Fourier spectrum. Our results establish diagnostics and benchmarks for upcoming quantum simulators of higher-dimensional LGTs.
Read the full abstract
Machine Learning
Constant regret in general games via higher-order optimism
We introduce an uncoupled learning algorithm which, when employed by all players of an arbitrary $N$-player normal form game with up to $K$ actions per player, guarantees $O(N^3\log^2 K)$ individual regret, uniformly over the horizon of play. The proposed algorithm - which we call higher-order optimism with discounting (HOOD) is a variant of optimistic follow-the-regularized-leader (OptFTRL) that combines a discounted $(N+1)$-th order predictor with entropic regularization over a suitable "lifting" of the game's strategy space. This combination of ingredients is purposefully designed to dampen large oscillations of the induced sequence of play in a controlled manner, removing in this way a key stumbling block of previous attempts to achieve constant regret in general games. Our approach bears several striking similarities to the concurrent - and completely independent - work of Liu, Farina, and Ozdaglar (arXiv:2608.31166), who very recently derived an $O(N^{21}\log^{4} K)$ regret bound through the use of higher-order optimism and an exponential moving average estimator.
Read the full abstract
Cosmology and Nongalactic Astrophysics
Consistency between cosmological and standard siren observations in evolving dark energy
A scalar field non-minimally coupled to gravity can be the driver of cosmic acceleration. Such a non-minimal coupling (NMC) can produce a non-zero gravitational-wave (GW) friction function $\alpha_M(z)$, which modifies the luminosity distance inferred from GW sources relative to its electromagnetic counterpart. We use a particular NMC scalar-tensor model, that explains time-varying dark energy in good alignment with DESI, to predict $\alpha_M(z)$ and the expected GW/EM luminosity-distance ratio $D_L^{\rm GW}/D_L^{\rm EM}$, and map it onto two common parametrizations—the $c_M$ and the $(\Xi_0,n)$ models. We find $c_M = -0.5\pm 0.2$ and $\Xi_0 = 0.88\pm 0.05$, $n=3.2\pm 0.3$, both consistent with GWTC-5 constraints at the $\lesssim 1\sigma$ level. This consistency is mostly driven by current large uncertainties in GW data, which lead to measurements consistent with both $\Lambda$CDM and the NMC model. By contrast, the dark energy constraints derived from analyzing cosmological data, under the common parametric scalar-tensor model using $\alpha_M(z)=c_M\Omega_\Lambda(z)/\Omega_{\Lambda 0}$ and the CPL parametrized equation of state $w_0w_a$, are in $2.2\sigma$ ($3.6\sigma$) tension with the NMC predictions for the $c_M$ ($w_0w_a$) parameter. We confirm that, in order to avoid cosmological instabilities, this parametrized model imposes strong implicit priors that are incompatible with physically-motivated scalar-tensor models when dark energy is dynamical.
Read the full abstract
General Relativity and Quantum Cosmology
Dyonic rotating cosmological black hole surrounded by quintessence
In this work, we constructed a new metric describing a rotating cosmological black hole with electric and magnetic charges in the presence of quintessence dark energy, where the cosmological constant and quintessence are associated with external matter. Starting from a Schwarzschild-type metric and considering all energy contributions in the function $f(r)$, we introduced rotation through the Newman–Janis algorithm using a unique and general complexification rule, avoiding possible ambiguities. Assuming that the metric satisfies the Einstein field equations with matter, we calculate the event horizon condition, the total stress–energy tensor $T_{\mu\nu}$, and the Kretschmann and Ricci scalars. We also analyze the influence of the cosmological constant $\Lambda$ and quintessence parameter $\alpha$ on the event horizon and ergosphere. In the limit $\alpha \rightarrow 0$, the scalar curvature differs from the vacuum result, $R \neq -4\Lambda$, while the singularity region remains unchanged. Finally, we investigate the angular and rotational velocities of a test particle, as well as the energy conditions, through the energy density and pressures required for the existence of this black hole solution.