Statistical Methodology
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Friday 4 September
We propose a Bayesian model-based approach for clustering functional data with an unknown number of clusters and within-curve correlated observations. Cluster-specific mean functions are represented using B-spline basis expansions, while within-curve dependence is modeled through an Ornstein–Uhlenbeck covariance structure. A truncated Dirichlet process mixture is used to infer the effective number of clusters, and a variational EM algorithm is developed for efficient posterior approximation. Simulation studies show that the proposed method performs well under both correctly specified and misspecified mean-function settings and achieves higher average values of the reported clustering metrics than the competing methods in the simulation settings considered. Comparisons with MCMC indicate that the variational approximation produces clustering results and parameter estimates that are in close agreement with those obtained by MCMC, while requiring substantially lower computational cost. An application to Canadian daily temperature curves further demonstrates the practical usefulness of the method in identifying interpretable functional clusters while accounting for within-curve dependence.
Machine Learning
KVMem: Virtualizing Million-Token Agent Workspaces on a Consumer GPU
Modern LLM agents operate in persistent workspaces whose accumulated history can exceed both GPU KV capacity and the model's native context window. Existing systems typically compact older context into summaries or retrieve it later as text, either losing fine-grained execution evidence or repeatedly prefilling content that the model has already processed. We present KVMem, a KV-context virtualization system that preserves overflowed workspace history as paged KV state across GPU memory, host memory, and NVMe. KVMem uses lightweight, model-native attention-space indexes to select relevant historical blocks and materializes a query-dependent execution view bounded by the model's native context window. Extensive evaluations on long-context agent benchmarks spanning histories up to one million tokens, including LongMemEval, MemoryAgentBench, and AgentLongBench, show that KVMem generally achieves higher task utility and greater inference efficiency than compaction-based approaches, the de facto standard for handling context overflow. In the DeepSWE long-context test with Qwen3.8-27B, KVMem improves task success from 43.8% with compaction-only context management to 48.4%. In our local-deployment evaluation, KVMem runs Qwen3.6/3.8-27B NVFP4 with MTP on an off-the-shelf laptop equipped with a 24\,GB RTX 5090 Laptop GPU, virtualizing agent workspaces of up to 1M tokens-four times the model's native 256K-token context window. In a single-session setting, KVMem generates $\sim$50 tokens/s, providing interactive responsiveness for local agent execution. More broadly, by decoupling addressable workspace size from the LLM's native context window, KVMem provides a practical path toward long-running agents whose workspaces can grow beyond that window.
Read the full abstract
Robotics
Coupled Control and Wireless World Models for Resilient Remote Robotic Control
Remote robotic systems operating over wireless networks must maintain reliable control despite limited communication resources, changing channel conditions, and environmental disturbances.However, continuously transmitting high-dimensional sensory observations, such as camera images, increases communication overhead and energy consumption while reducing robustness under unreliable connectivity.To address these challenges, this paper proposes a resilient communication-aware remote robotic control framework based on coupled control and wireless Joint Embedding Predictive Architecture (JEPA) world models that jointly capture robot dynamics and wireless channel evolution from visual observations and a combination of raw and structured radio frequency (RF) representations based on spectrograms and Persistence Images(PIs).The learned latent representations enable predictive communication scheduling by jointly forecasting future robot states and wireless conditions, thereby reducing unnecessary uplink transmissions while maintaining reliable control performance.Furthermore, an adaptive resilience mechanism detects latent prediction discrepancies and efficiently adapts perception embeddings to accommodate wireless and visual environmental changes without retraining the complete control policy.The proposed framework is evaluated in a synchronized Gazebo-Robot Operating System (ROS)-Sionna robot-wireless simulation environment under diverse wireless propagation and perception perturbations.Experimental results demonstrate significant improvements in communication efficiency, robustness, and resilience while maintaining navigation performance compared with conventional Proportional Integral Derivative (PID), model-free Deep Q-Network (DQN), and predictive approaches based on Vision Transformers(ViTs).
Read the full abstract
Artificial Intelligence
ElderBench: Benchmarking Autonomous Mobile Agents for Older Adults
While autonomous mobile agents hold great potential for assisting older adults with smartphone usage, existing GUI benchmarks mainly rely on explicit, goal-oriented instructions and rarely capture the naturally occurring language patterns of older users, such as indirect speech, referential ambiguity, and under-specified requests. This mismatch between benchmark instructions and real-world elderly interactions may hinder reliable agent deployment. To address this gap, we present ElderBench, the first benchmark for evaluating mobile GUI agents in authentic elderly-oriented scenarios. ElderBench is constructed from 249 naturally elicited smartphone tasks collected from older adults across 20 applications. We first characterize the linguistic divergence between elderly instructions and existing GUI benchmark instructions from syntactic, semantic, and pragmatic perspectives. We then evaluate mainstream GUI agents and Vision-Language Models under both online and offline settings, revealing substantial performance degradation when handling elderly-oriented instructions. Through controlled instruction normalization, failure analysis, and fine-grained linguistic feature analysis, we further identify how elderly-specific language patterns contribute to agent failures. Our findings provide actionable design insights toward more adaptive, interpretable, and age-inclusive GUI agents for older adults.
Read the full abstract
Probability
Recombination in discrete and continuous time from the viewpoint of Markov embedding
The classic recombination equation, both in discrete and in continuous time, can be solved in a way that derives from the Markov chain of a partitioning process. Here, we revisit this structure from the point of view of the Markov embedding problem. In particular, we analyse when a discrete-time Markov matrix of recombination type can occur in a time-homogeneous Markov semigroup that is generated by a recombination rate matrix. En route, we also show that such rate matrices (or Markov generators) generally do not form a matrix algebra, but span a real Lie algebra.
Read the full abstract
Signal Processing
3D Radar Imaging from the UAV Nadir
Radars improve the sensing robustness of UAVs by operating under poor lighting and weather conditions and seeing through occlusions such as vegetation. However, they suffer from poor angular resolution, which can be addressed using synthetic aperture radar (SAR) algorithms. State-of-the-art UAV SAR methods operate at a depression angle and are not suitable for sensor fusion applications where the data are collected from areas directly below the UAV (i.e., the UAV nadir). In this paper, we present an interferometric SAR (InSAR) framework for reconstructing 3D images from the UAV nadir using a low-cost multi-input-multi-output (MIMO) mm-wave radar. Additionally, an effective method based on the phase gradient autofocus (PGA) is presented for compensating the phase error across the virtual receive antennas. We demonstrate the effectiveness of our 3D imaging algorithm in both simulation and experimental scenarios.
Read the full abstract
Atomic Physics
Dead-zone-free free-induction-decay alkali-metal atomic magnetometer
The detection dead zone is an important systematic limitation in scalar atomic magnetometers, constraining their practical utility. In this work, we demonstrate a sensitive dead zone free scalar magnetometer by integrating previously established techniques into a FID magnetometer based on Bell Bloom optical pumping. The dead zone is eliminated by inserting a reflecting mirror within a multipass cavity assisted atomic cell, which folds the optical beam into orthogonal paths. Our analysis reveals that interregion cross talk is essential for interpreting the resulted experimental signals. The sensor exhibits oscillation signal amplitude variations within a factor of three across all orientations in three dimensional space, and a field sensitivity better than 80 fT/Hz^(1/2) over the full space. We further characterize the heading error in the geomagnetic field range, where experimental results agree with theoretical predictions within 0.7 nT. Additionally, the sensor can operate in a closed loop mode by feeding back the real time frequency extracted from the FID signal to modulate the pump beam, we characterize the magnetic field slew rate of the sensor. This work paves the way towards a sensitive vector FID magnetometer.
Read the full abstract
Computer Vision
LetOccVote: Learning Weakly Supervised 3D Occupancy through Consensus
Weakly supervised 3D occupancy prediction reduces the reliance on costly 3D annotations by learning from 2D pseudo-labels generated by vision foundation models. However, existing methods typically use these imperfect pseudo-labels directly as supervision, making occupancy learning vulnerable to erroneous geometric and semantic targets. We observe that agreement across repeated observations provides an inexpensive and reliable cue for assessing pseudo-label reliability. Based on this observation, we propose LetOccVote, a weakly supervised Gaussian-based occupancy framework that leverages cross-frame voting to improve both geometric and semantic supervision. For geometry, Depth Vote exploits cross-frame geometric agreement to refine supported pseudo depth and reject contradictory estimates before volumetric lifting and depth supervision. For semantics, Semantic Vote aggregates pseudo-semantic observations in a shared 3D space to identify reliable and contested evidence, strengthening reliable semantic supervision while filtering unreliable pseudo-label segments. The entire framework is trained solely with 2D pseudo-label supervision without requiring 3D occupancy annotations. On Occ3D-nuScenes, LetOccVote achieves 53.27 IoU and 20.39 mIoU, establishing state-of-the-art performance among methods with 2D pseudo-label supervision.
Read the full abstract
Materials Science
Multi-task deep-learning optimization of trade-off properties for superior-performance Fe-based soft magnetic alloys
Fe-based amorphous alloys are promising soft magnetic materials for developing next-generation devices with high frequency and efficiency. However, optimization of Fe-based alloys with ultra-high saturation magnetic flux density (B_s), ultra-low coercivity (H_c), and good glass-forming ability is a notorious problem, owing to the vast composition space and complex trade-offs of these properties. Thus, conventional design methods encounter great challenges. Here we develop a generative multi-task deep learning (GMTDL) to achieve simultaneous optimization of compositions and trade-off properties. The GMTDL can sufficiently exploit and share the knowledge of datasets across different tasks, despite the limitation and imbalance of these datasets. Therefore, it exhibits superior performance in prediction of alloys with multiple targeted properties, outperforming previous machine learning-based design strategies. Moreover, the GMTDL can also tailor compositions, providing an efficient way to regulate properties and generate desired candidates for further experimental processing. The validity and reliability of GMTDL are rigorously tested by benchmarking with Fe-based alloys reported very recently. Moreover, some new alloys with ultra-high B_s and ultra-low H_c are predicted. The optimal content windows of key elements and their synergistic effects are also unraveled for practical guidance. Thus, our study establishes an effective and reliable paradigm for simultaneous prediction and optimization of high-performance materials with multiple properties.
Read the full abstract
High Energy Physics (Theory)
What is Classical Null String Theory?
We revisit the classical definition of a null string in a D-dimensional Minkowski target space and distinguish a consistent partially-gauged Carrollian sigma-model from the null string theory. The Isberg-Lindström-Sundborg-Theodoridis (ILST) action gauges worldsheet diffeomorphisms and the common Weyl rescaling, but not the independent relative rescaling of the temporal and spatial representatives of the intrinsic Carrollian structure, Carroll-Weyl scaling. A physical string theory requires every local change of worldsheet representative to be gauged, therefore, the Carroll-Weyl symmetry should be gauged in a null string theory. In the minimal realization, the corresponding gauging is obstructed at finite tension, whereas at zero tension it yields an additional first-class constraint and the associated identification of physical configurations along its gauge orbits. The description of the null worldsheet as a congruence of null geodesics provides a complementary target-space interpretation of gauging of the Carroll-Weyl scaling. Our analysis thus provides the supplementary requirements that turns the ILST theory into the null string theory.
Read the full abstract
Differential Geometry
Connection matrices and Berry phases on the homogeneous spaces $\mathcal{D}^J_1$ and $\mathcal{X}^J_1$ attached to Jacobi groups
We analyze the relation between connection matrices and Berry phases on the Siegel-Jacobi disk $\mathcal{D}^J_1$ and Siegel-Jacobi upper half-plane $\mathcal{X}^J_1$ and we explicitly compute the connection matrices and the Berry connections on these homogeneous spaces in different coordinates.
Read the full abstract
Computation and Language
MMTClinic: Multimodal, Multilingual Time Series Question Answering and Reasoning Benchmark for Clinical Domain
Time-series data in clinical settings is crucial for capturing dynamic changes in a patient's health over time, enabling timely diagnosis, personalized treatment, and early detection of critical events. However, the development of clinically reliable and linguistically inclusive medical AI systems remains a significant challenge, primarily due to the lack of multimodal, multilingual, and time-series-grounded benchmarks that reflect the complexity of real-world clinical scenarios. To fill this gap, we present MMTClinic, a benchmark designed to evaluate large language models (LLMs) on complex reasoning and question-answering tasks involving clinical time-series. MMTClinic combines text, medical images, and multivariate physiological signals and includes 30,000 QA pairs (15,000 multiple choice questions (MCQs) and 15,000 open-ended questions) across five languages: English, Hindi, Bengali, Marathi, and Tamil. These questions cover three important clinical tasks—mortality prediction, heart rate forecasting, and SOFA score estimation. We evaluate 13 state-of-the-art LLMs in zero-shot, few-shot, and chain-of-thought settings. Our evaluation reveals notable differences in model performance across tasks, languages, and modalities, highlighting current limitations in clinical reasoning capabilities. MMTClinic provides a valuable resource for advancing multilingual, multimodal, and time-series-aware medical AI research. The dataset will be made publicly available on successful acceptance of the work.
Read the full abstract
Computation and Language
MABPD: Multi-Agent Bias Probing & Detection via Structured Argument Debate
Media bias in news articles operates through subtle linguistic cues—loaded language, selective framing, and strategic omission—that resist single-model detection and have traditionally required large annotated corpora for supervised training. We ask whether structured multi-agent deliberation can serve as a principled, training-free alternative to supervised classification for this task. We introduce MABPD (Multi-Agent Bias Probing & Detection), a pipeline in which three specialized LLM agents analyze an article from complementary perspectives and resolve disagreements through a Structured Argument Debate (SAD) protocol. SAD implements a domain-motivated asymmetric burden of proof—biased claims without grounded textual evidence carry zero weight—combined with role-weighted voting and post-consensus verification, replacing task-specific supervised decision boundaries with explicit deliberative structure. Ablation confirms that this structured deliberation, not mere agent parallelism, drives performance: removing the debate module reduces F1 by up to 10.6 points. On the BABE benchmark (4,121 expert-annotated sentences), MABPD achieves 83.4% macro F1 on the held-out test split—within 0.7 percentage points (pp) of the supervised SOTA (MAGPIE, 84.1% macro F1; Horych et al., 2024)—without any task-specific training or threshold tuning on annotated data. Cross-dataset evaluation on the SemEval 2019 HyperPartisan corpus (644 articles) yields 75.0% zero-shot accuracy, within 7.2 pp of the supervised SOTA accuracy (82.2%; Kiesel et al. 2019), confirming transfer across annotation regimes. We release the full pipeline and evaluation code.
Read the full abstract
Artificial Intelligence
Long Horizon Transformer Quantile Fault Prediction for Multi Site Industrial Predictive Maintenance
Long-horizon predictive maintenance requires models to distinguish slowly evolving degradation from normal operating-regime variation over planning windows measured in days rather than hours. This paper evaluates whether an explicit conditional-quantile representation provides an informative classifier interface for this problem. The proposed TQRNN30d framework combines a dual-stage quantile regression neural network (QRNN) feature extractor with a multi-stream temporal fusion classifier. Each hourly word of 81-channel machine behaviour is mapped to a 324-dimensional quantile-state representation, and 720 ordered hourly words form the 30-day document supplied to the long-horizon model. The classifier fuses quantile states with dynamic covariates, channel-level static metadata, and a 168-hour latent-history stream using gated residual processing, causal recurrent encoding, and metadata-conditioned cross-modal attention. A bounded instability-aware signal derived from sustained one-word-ahead prediction-error divergence provides auxiliary memory modulation at the longest horizon. Evaluation uses a machine-disjoint 43/14/15 train/validation/test allocation across 72 machines in nine manufacturing facilities. At 30 days, TQRNN30d achieves 79.97% F1, 80.18% recall, 81.82% precision, 82.39% accuracy, and 0.820 ROC-AUC. It leads all 18 evaluated baselines at the 7-, 14-, and 30-day fixed-threshold comparisons, with the largest F1 advantage at 14 days. The results support held-out-machine performance within the observed homogeneous nine-facility fleet, but do not establish unseen-site, cross-equipment, or cross-sector generalisation.
Read the full abstract
High Energy Physics (Theory)
Form Factor Alphabets and Antipodal Duality from Cluster Algebras
We develop a cluster-algebraic description of symbol alphabets for chiral stress-tensor form factors in planar $\mathcal N=4$ super-Yang–Mills theory. Motivated by unfolding the three-point $C_2$ quiver, we propose a periodic framework for $n$-point form-factor alphabets based on folding a $\mathrm{Gr}(4,2n)$ quiver with two periods of frozen nodes. At four points, the folding leads to the $\widetilde{\mathrm{VI}}$ cluster algebra, which has infinitely many cluster variables. Tropical truncation selects a finite set of rational coordinates and four physical limit rays. The $A_1^{(1)}$ mutation sequences approaching these rays generate the four square roots and the corresponding algebraic-letter spaces. Within the resulting finite set of rational candidates, antipodal closure excludes precisely eight additional letters. On the parity-preserving surface, the antipodal map is realized, after a further folding, by two commuting mutations followed by a relabelling. Furthermore, we find an alternative quiver within the same mutation class that becomes self-antipodal after parity folding, mirroring the antipodal self-duality of the four-point MHV form factor. Its double- and triple-collinear boundaries reproduce, respectively, the $C_2$ algebra of the three-point form factor and the $A_3$ algebra of the six-point amplitude; after folding the latter to $C_2$, the antipodal map exchanges the two $C_2$ boundaries.
Read the full abstract
Nuclear Theory
Microscopic Investigation of $\alpha$ Clustering for the $\alpha$ Decay Based on the Density-Constrained Frozen Hartree-Fock Calculations
The inter-nucleus potential between an $\alpha$ particle and a daughter nucleus ($^{100}$Sn) is calculated based on the density-constraint frozen Hartree-Fock (DCFHF) method with a Skyrme-type energy density functional (EDF). It is shown that the potential has a repulsive core at short distances with an attractive pocket close to the nuclear surface. We have confirmed that the potential obtained with the Skyrme DCFHF method agrees well with that obtained with the extended Thomas-Fermi approximation to the Skyrme EDF. Furthermore, it is demonstrated that the height of the repulsive core decreases by taking into account melting of the $\alpha$ cluster inside a nucleus. With the aid of the nucleon localization function, we show that the obtained state in DCFHF clearly manifests localization of spin-up and spin-down neutrons and protons, suggesting the possible emergence of an $\alpha$-cluster-like structure in the DCFHF wave function. These results opens up a possibility to investigate $\alpha$ clustering in heavy nuclei within mean-field calculations.
Read the full abstract
Computer Vision
PAPT++: Risk-Aware Adversarial Tuning and Generation for Single Domain Generalization
Single domain generalization (SDG) aims to learn a model from one labeled source domain that generalizes to unseen target domains. A common strategy is to enrich the source distribution with augmented or generated samples, and recent text-to-image (T2I) diffusion models provide a strong generative prior for this purpose. However, diversity alone is insufficient for robust generalization, because useful generated samples should also capture variations that the current classifier finds difficult. Motivated by distributionally robust optimization (DRO), we define a semantic ambiguity set in the class-conditional generative space of a pretrained T2I model and search it for samples with high classification loss under the current classifier. To this end, we introduce PAPT++, a risk-aware adversarial generation-training framework for SDG. PAPT++ first learns diverse semantic reference images for each class through image-text alignment and intra-class diversity regularization. These references then serve as denoising targets during classifier-guided diffusion synthesis, reducing semantic drift while guiding generation toward challenging variations. The generated samples are combined with the source data to update the classifier, and the updated classifier guides the next synthesis round in return. In this way, PAPT++ progressively exposes the classifier to challenging yet semantically consistent variations. Extensive experiments on standard SDG benchmarks demonstrate the superiority of the proposed PAPT++ method and the effectiveness of its main components.
Read the full abstract
Signal Processing
Hierarchical Codebook Design and Low-Overhead Beam Training for Near-Field Communications With Uniform Circular Arrays
Extremely large-scale multiple-input multiple-output (XL-MIMO) enables near-field location-specific beam focusing for sixth-generation (6G) communications. Uniform circular arrays (UCAs), with rotational symmetry and uniform azimuth coverage, have emerged as a key enabling architecture for near-field XL-MIMO systems. In this paper, we propose a resolution-aware hierarchical codebook for near-field UCA systems, along with an efficient two-stage beam training scheme to significantly reduce the training overhead. Specifically, we characterize the minimum resolvable distance of UCA systems in the near-field region based on a geometric spherical-wave propagation model, revealing their spatial resolution capability in the joint angle–distance domain. Guided by this result, we design a UCA-specific hierarchical codebook, where a power-efficient distance-robust beamforming (DRBF) codebook provides coarse azimuth localization and a full-precision (FP) codebook sampled according to the minimum resolvable distance enables refined angle–distance beam search. The regularized modal compensation suppresses weak-mode amplification and provides a controllable tradeoff between absolute amplitude gain under unit-norm transmission and distance robustness. Based on this codebook, we develop a hierarchical decoupled-architecture Bayesian regression (HDA-BAR) scheme for fast and accurate near-field beam training. For the considered array configuration, the resulting HDA-BAR training procedure requires $384$ probing slots, corresponding to an approximately \(99.66\%\) overhead reduction relative to the conventional near-field exhaustive-search benchmark.
Read the full abstract
Computation and Language
On Epistemic Diversity in Large Language Models
Large language models (LLMs) are increasingly used not only to retrieve information, but to answer questions, explain, teach, and support inquiry. In such settings, evaluation cannot be exhausted by accuracy or alignment alone. A system may give a correct answer while still narrowing users' access %to knowledge. to alternative valid answers, explanations, or reasoning routes. Drawing on the broader notion of epistemic diversity in philosophy and social epistemology, we formalize it in the context of LLMs as the range of valid answers, explanations, and reasoning routes that an LLM exposes to users. We argue that epistemic diversity is a useful evaluation dimension for settings where LLMs are used to support knowledge-intensive tasks. We propose a preliminary framework for conceptualizing and measuring epistemic diversity in LLMs, and operationalize it in two domains. We find that frontier LLMs often exhibit epistemic narrowness, repeatedly collapsing large valid answer spaces onto small canonical subsets. These findings suggest that LLM evaluation should move beyond accuracy-oriented paradigms and treat epistemic diversity as an important dimension of model capability.
Read the full abstract
Probability
Eco-evolutionary cycles in a matching type predator-prey interaction
We study the population dynamics of a predator-prey system with two types in each species. Within a species, predator or prey, dynamics are described by a neutral competitive Lotka-Volterra model, i.e., birth, death and competition parameters are equal for both types. Additionally, we assume that the intra- and inter-type competition parameters are equal. The predator-prey interaction is defined by a matching-types model where predators of type $i$ exclusively interact with prey of type $i$. The individual-based model is described by a birth-death process with immigration, where immigration reflects mutations between the types of the same species. We completely describe the deterministic dynamics arising as a large population limit of this birth-death process. We find that depending on the parameters, potential equilibria are the coexistence of all four types, coexistence of a non-matching or matching pair of predators and prey, or the extinction of the predator or prey species resulting in a line of two-type equilibria. When mutations are sufficiently rare, then the predator-prey dynamics are described by successive jumps between the different deterministic equilibria on this mutational time scale. These jumps describe eco-evolutionary cycles of repeated prey or predator invasions and declines. When coexistence of all the types is possible, we show that these cycles accumulate on this time scale. Lastly, to prove that after the accumulation point the system converges to the coexistence equilibrium, we consider a slightly modified model with unequal intra- and inter-type competition parameters. This modified setting allows us to conclude that after the accumulation point all four populations remain macroscopic and converge to the coexistence equilibrium.
Read the full abstract
Complex Variables
Norm of the generalized Hilbert operator on weighted Bergman spaces
Several upper bounds as well as one lower bound for the operator norm of the generalized Hilbert operator $\mathcal{H}_b$ acting on weighted Bergman spaces $A_{\alpha}^p$ are established. Moreover, under some mild assumptions, we obtain the exact norm of $\mathcal{H}_b$ on $ A_{\alpha}^p$.
Read the full abstract
Machine Learning
PACE: Propagation-Aware Collaborative Correction for One-Shot Personalized Federated Graph Learning
Client heterogeneity creates both an opportunity and a risk in personalized federated graph learning. Knowledge held by other subgraphs may complement a receiver's Local model, but an incompatible transfer can override reliable predictions. One-shot communication sharpens this tension because an unsuitable server return cannot be corrected later. We introduce PACE, which treats collaborative knowledge as a compact correction to a complete Local predictor rather than as its replacement. Each client uploads a rank-r update carrier and a diagonal sketch of propagated message moments. The server uses them to construct a propagation-aware, receiver-anchored correction, while the receiver retains its full Local model. Convex negative-log-likelihood calibration (CNLL) then selects one coefficient between Local and External logits using validation nodes; model parameters remain fixed and no feedback is sent. At Rank-6, personalized returns occupy 9.6-17.6% of dense tensor bytes across the six evaluated datasets. The correction receives nonzero weight and improves both Accuracy and weighted-F1 over Local on five datasets; on ogbn-arxiv, CNLL assigns zero predictive weight to the correction and preserves Local predictions exactly. Applying the same CNLL rule to matched baselines on three citation datasets does not account for these gains. The central result is therefore that a small transported correction can augment a complete Local model when receiver evidence supports it while leaving the Local prediction unchanged otherwise.
Read the full abstract
Commutative Algebra
On the closed neighborhood ideal of the square of broom and double broom graphs
Classes of squarefree monomial ideals were intensively studied being an important connection between two main areas in mathematics: commutative algebra and combinatorics. Many algebraic invariants of the squarefree monomial ideals are given in terms of properties of the combinatorial associated objects (graphs, simplicial complexes etc.).
Read the full abstract
Machine Learning
Communication-Efficient Personalized Federated Learning via Layer-Wise Multi-Threshold Random Sketching
Personalized federated learning (PFL) is a promising paradigm for collaborative learning over distributed devices, where edge nodes collaboratively train personalized models without sharing raw data. Although PFL addresses data heterogeneity by learning client-specific models, it still suffers from substantial uplink and downlink communication costs when exchanging high-dimensional parameters in bandwidth-constrained systems. Recent one-bit methods achieve extreme compression, but they usually rely on a single thresholding rule applied to the whole model. This design has two limitations. First, it overlooks layer-wise differences in parameter distributions and quantization sensitivities. Second, a single threshold provides only coarse binary information and cannot capture fine-grained variations in parameter distributions. To address these issues, we propose a communication-efficient PFL framework via layer-wise multi-threshold random sketching. In the proposed method, each layer is assigned its own set of quantization thresholds, so that the compressed representation can adapt to layer-specific statistics while using multiple intervals to provide a finer low-bit description of sketched parameters. The proposed method supports bidirectional communication using compact low-bit sketches and improves the communication-accuracy tradeoff compared with existing one-bit compression approaches.
Read the full abstract
Cosmology and Nongalactic Astrophysics
A guide to choosing data compression methods for cosmological inference
We provide a pedagogical guide to help researchers choose an appropriate compression method for cosmological inference and numerical covariance estimation, trading off the balance between information loss and complexity. We describe methods used in the literature, categorising them by the form of the loss function – Fisher information, mutual information, or mean squared error – and by whether they are linear or non-linear. We consider in detail the linear compression methods: massively optimised parameter estimation and data compression (MOPED), canonical correlation analysis (CCA), principal component analysis, and linear neural networks. We also investigate simple non-linear methods based on neural networks with similar architectures but different loss functions. We use tomographic weak lensing power spectra as example data vectors, deriving constraints on the cosmological parameters $\Omega_\mathrm{m}$, $\sigma_8$, $w_0$ and $w_a$ from compressed data. For our figure of merit (FoM) we use the log determinant of the Fisher matrix, which approximates the covariance of the full posterior well. MOPED and a Fisher information-based linear neural network both achieve lossless compression. Neural networks with loss functions which optimise mutual information achieve at most 84 per cent of the uncompressed FoM and those which minimise mean squared errors achieve at most 72 per cent. We show how to determine whether MOPED will retain information even if the fiducial cosmology is not known with certainty, and demonstrate a strategy for CCA which results in an FoM close to that of MOPED while requiring only an approximate idea of where the bulk of the posterior mass is located.
Read the full abstract
Sound
ProLombard: Structured Multi-Scale Modeling for Normal-to-Lombard Speech Conversion
Normal-to-Lombard (N2L) speech conversion aims to improve speech intelligibility in noisy environments by transforming normal speech into Lombard-style speech while preserving linguistic content, speaker identity, and speech quality. Despite recent progress, existing methods typically model the Lombard effect at the utterance level or the frame level, overlooking its hierarchical nature and its entanglement with both speaker identity and phoneme-level content. This limitation leads to Lombard leakage in speaker representations and incomplete separation between Lombard characteristics and linguistic content. In this work, we propose ProLombard, a structured multi-scale N2L framework that explicitly models the Lombard effect across utterance-, phoneme-, and frame-level representations. To address Lombard-speaker entanglement, we introduce an aligned speaker encoder (ASE) that suppresses Lombard leakage by aligning Lombard-speech speaker embeddings with their normal-speech counterparts. To achieve more complete Lombard-content disentanglement, we develop a phoneme-aware disentanglement and injection mechanism that extends conventional frame-level modeling to the phoneme level. Furthermore, we design a vector quantization (VQ)-median module that provides robust phoneme-level representations through VQ-based segmentation and median-frame-based aggregation. Extensive experiments on Mandarin and English Lombard datasets demonstrate that the proposed approach consistently improves speech intelligibility, Lombard similarity, and perceptual quality over baselines while maintaining speaker identity. These results highlight the importance of structured multi-scale modeling for effective N2L speech conversion.
Read the full abstract
Computer Vision
Weather-Conditioned Depth Anything
Monocular depth estimation foundation models, such as the Depth Anything series, have achieved remarkable performance across diverse domains. However, they still suffer from critical failures under adverse weather conditions, such as fog, rain, snow, or at night. To address this, we present Weather-Conditioned Depth Anything (DA-W), a framework that explicitly disentangles style from content for weather-robust depth estimation. Specifically, we introduce a Style Filter trained on a curated mix of real and synthetic degradation datasets to extract content-independent, degradation-aware weather embeddings. This style embedding is then injected into the Depth Anything backbone using a parameter-efficient, zero-initialized adapter. Such a lightweight modulation allows a single unified model to robustly adapt to diverse conditions, including fog, rain, snow, and low-light, while avoiding catastrophic forgetting of its core generalization abilities in normal conditions. We train the adapter using a pseudo-label distillation and alignment strategy. Our comprehensive experiments demonstrate that our proposed DA-W achieves state-of-the-art robust depth estimation, improving AbsRel by an average of 3.7% on our curated weather benchmarks, while matching or slightly outperforming performance on standard clean benchmarks. Our project page is available at https://zhaoming-tamu.github.io/WCDA/.
Read the full abstract
Probability
Stochastic fractional diffusion equations with spatial Gaussian noise in the Stratonovich regime
For a class of stochastic fractional diffusion equations, we prove the existence of Stratonovich solutions and obtain the precise values of the Lyapunov exponents governing the large $t$-asymptotics of their $p$-th moments for all real $p\in\{1\}\cup[2,\infty)$, as well as the large $p$-asymptotics of the $p$-th moments.
Read the full abstract
Data Structures and Algorithms
Beyond Distance Ordering: Resource Complexity and Universal Optimality of Exact Labeled Directed Shortest Paths
We study exact single-source shortest paths when the output is only the materialized labeled distance vector ($\mathrm{DIST}$), rather than a distance order. In the full deterministic comparison-addition model, the minimum worst-case number of additions on every fixed directed topology is exactly the maximum number $\rho_{\mathrm{fwd}}$ of forward nonsource endpoint classes over rooted vertex orders; the lower bound permits adaptive control, literals, and arbitrary mixed sums. This arithmetic law aligns with the comparison optimum on DAGs, where the full resource region is an exact rectangle. Cycles destroy that alignment: a two-spoke shared-hub graph has coordinatewise optima $(4,2)$ but requires five comparisons at the two-addition budget. Its $k$-spoke extension forces $k\log_2 k+O(k)$ comparisons at the addition optimum and has an entropy-tight deterministic tradeoff $C_{k+r}^*(H_k)=\Theta(k+\Lambda_{k,r})$, where $\Lambda_{k,r}=\log_2(k!/[r!(r+1)^{k-r}])$, with leading constant one when $\Lambda_{k,r}/k\to\infty$. Because the two coordinatewise minima need not belong to one program, these conflicts lead to the same-program benchmark $\operatorname{OPT}_{\mathrm{DIST}}=\inf_A\sup_w(C_A(w)+P_A(w))$. An exact transcript-cone game yields one uniform interpreter whose charged addition-comparison cost equals $\operatorname{OPT}_{\mathrm{DIST}}$ on every topology; its optimal actions are synthesizable in polynomial space but may require exponential time. Finally, an active-core reduction and the current deterministic directed-SSSP bound give an efficient uniform $O\!\bigl(\operatorname{OPT}_{\mathrm{DIST}}\sqrt{\log(2+\operatorname{OPT}_{\mathrm{DIST}})\log\log(4+\operatorname{OPT}_{\mathrm{DIST}})}\bigr)$ charged-operation bound. Thus optimal numerical policies exist uniformly, while efficient constant-competitive navigation remains open.
Read the full abstract
Computation and Language
Generating Constructive Feedback on Stories via Reinforcement Learning
Constructive feedback is crucial for creative writers to refine their storytelling abilities. Since receiving feedback from human experts is often costly and time-intensive, large language models (LLMs) offer a scalable and efficient alternative as automatic writing assistants. Despite their potential, research indicates that LLM-generated feedback is often generic, lacks actionability, and fails to identify which writing issue is most critical. To address these limitations, we present a reinforcement learning approach that steers LLMs to generate constructive feedback without the need for ground-truth feedback. We train our model using group relative policy optimization (GRPO) with a novel multi-component reward function aiming at constructiveness: it prioritizes feedback that is uniquely tailored to the story, helps to improve story quality, and addresses the most critical writing issue. In automatic and human evaluation across three story corpora, our approach outperforms state-of-the-art LLMs (including Gemini) and competitive baselines. We find that providing actionable suggestions is the main driver of feedback constructiveness.