Peer Revenue Past citations are no guarantee of future returns

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Sunday 13 September

Distributed Computing

Joint Optimization for Federated Learning and Transmission over Unreliable Wireless Networks with Heterogeneous Data

In wireless federated learning (FL), data heterogeneity and multiple local updates induce client drift, degrading model convergence. It is further affected by unreliable wireless links, as transmission errors may invalidate model updates. To address these challenges, we propose a federated random walk averaging (FedRW) framework, which is a variant of federated averaging (FedAvg) that mitigates data heterogeneity by updating models along random walk (RW) paths and aggregating them at the server. Model parameters are transmitted in packets with retransmission support to improve training quality by mitigating wireless errors along RW paths. Meanwhile, wireless transmission delays hinder the exploration of FedRW. To this end, we formulate a joint optimization problem that integrates learning, RW path selection, and transmission parameter tuning, aiming to minimize the training loss under delay constraints. By deriving an upper bound on the expected convergence of FedRW over unreliable wireless networks, we reduce the problem to a general form agnostic to task type and model architecture. A distributed solution is then proposed, in which the server or clients optimize packet size and maximum number of retransmissions locally, and efficiently select reliable and expandable next-hop nodes via a resilience-aware beam search with dynamic pruning. Simulation results show that FedRW achieves 2.26%-9% higher accuracy than state-of-the-art baselines under high data heterogeneity. Furthermore, the jointly optimized FedRW yields at least 2.78% higher accuracy and faster convergence compared to baselines.
Read the full abstract

All authors Changheng Wang, Xianchao Zhang, Zhiqing Wei, Lingzhu Zhao, Zhongming Yang, Zhiyong Feng

Computation and Language

The Attribution-Compression Frontier in Retrieval-Augmented Generation

Context compression reduces generator input in retrieval-augmented generation, but answer quality alone does not characterize citation attribution. We measure citation attribution across compression methods and budgets, comparing reranking, extractive selection, abstractive summarization, token pruning, and an extract-cluster-rewrite construction on ASQA and QASPER under a fixed generator and primary entailment evaluator. On ASQA at a nominal 0.25 budget (achieved compression 0.08), a RECOMP-style compressor's citations score 0.86 precision against its summaries but 0.12 against source spans under our re-attributability protocol. These estimates depend on a shared NLI model for span recovery and citation scoring and lack independent human calibration. Extractive selection's observed grounded precision ranges from 0.43 to 0.49 across nominal budgets from one-half to one-tenth of the ASQA context, while answer quality declines. For the same RECOMP setting, claim verification after source recovery yields an unsupported rate of 0.88 versus 0.17 when checking summaries. This gap persists beyond structural rejection of missing provenance, but remains evaluator-dependent. A 200-question TRUE T5-XXL audit also finds emitted–grounded gaps under both fixed and recomputed source mappings, without establishing human-calibrated support rates.
Read the full abstract

Analysis of PDEs

Global Stability of 3D Compressible Non-Resistive MHD: Hidden Damping and Rational Background Fields

We prove the global well-posedness and nonlinear stability of classical solutions to the three-dimensional compressible viscous, non-resistive MHD system on $\mathbb{T}^3$ near the equilibrium $(1,\mathbf{0},\mathbf{e}_3)$, for small $x_3$-symmetric perturbations and without any Diophantine condition on the background magnetic field. The central obstruction is the $x_3$-independent sector, in which the density and the magnetic field possess no dissipation, no damping, and no decay. We overcome it by exhibiting a hidden wave structure for the exact total pressure $\mathcal{D}=P(1+a)+B_3+\frac12|\mathbf{B}|^2$, the averaged pair $(\overline{\mathbf{u}},\overline{\mathcal{D}})$ obeys a closed system, and $\overline{\mathcal{D}}$ satisfies a strongly damped wave equation whose principal wave part propagates at the fast magnetosonic speed and whose non-parabolic branch damps at a rate that involves no gain of derivatives. This hidden damping substitutes for the missing magnetic dissipation, and simultaneously absorbs the magnetic pressure $\frac12\nabla|\mathbf{B}|^2$, the main obstruction created by compressibility. Together with a damped wave structure for the oscillatory sector and space-time weighted energy functionals with shifted time weights, this yields global existence, uniform stability, and explicit polynomial decay rates.
Read the full abstract

All authors Lin-An Li, Qian Li, Jiahong Wu, Xiaojing Xu

Data Structures and Algorithms

A $(p+q)^{O(pq)}$-approximation for $(p, q)$-Flexible Graph Connectivity

In the $(p,q)$-Flexible Graph Connectivity problem, the input consists of non-negative integers $p$ and $q$ and a graph $G=(V, E)$ whose edges are classified into safe and unsafe edges with non-negative edge costs. A subgraph H of G is $(p,q)$-Flex-Connected if every non-empty proper subset of vertices has either at least $p$ safe edges or at least $p+q$ total edges crossing it. The goal is to find a minimum cost subset $F\subseteq E$ of edges such that the subgraph $(V, F)$ is $(p,q)$-Flex-Connected. We give a $(p+q)^{O(pq)}$-approximation for this problem, which in particular implies a constant approximation for every fixed constants $p$ and $q$. We achieve this by designing a $(p+q)^{O(pq)}$-approximation for the augmentation problem of finding a minimum cost subset of edges to add to make a (p,q-1)-Flex-Connected graph into a (p,q)-Flex-Connected graph. Underlying the augmentation algorithm is a structural result showing that all deficient cuts can be represented by min rooted-cuts in a $(p+q)^{pq}$-sized collection of digraphs. This structural result was discovered by ChatGPT Astra.
Read the full abstract

Mathematical Physics

Liouville integrability of Beltrami-Maxwell fields with sources: A geometric theory of magnetic surfaces

We study classical Maxwell fields with non-trivial sources in four-dimensional Minkowski spacetime. In particular, contact and symplectic geometric aspects are discussed. As a point of departure, it is shown that a few classes of solutions with sources are constructed from the so-called Beltrami fields, where these fields are contact forms on three-dimensional Riemannian manifolds. In addition, several applications in these classes are provided. One is to give a solution to the London equations for superconductors. Another one is that, if there is a non-trivial conserved quantity along the magnetic vector field, then the Maxwell system can be viewed as a Liouville integrable Hamiltonian system on a four-dimensional symplectic manifold, where this manifold is obtained by a symplectization of the contact manifold. As examples, Maxwell fields with the Lundquist model, a restricted case of the ABC flow model, and so on, are shown to be integrable in the above sense.
Read the full abstract

Mesoscale and Nanoscale Physics

The effect of light scattering in cavity electrodynamics: Fresnel equations with decoherence

We consider the effect of light decoherence in a Fabri-Perot (FP) microcavity. We show that, when present, it may significantly change its linear response to the incident light.The optical properties of the bounding mirrors are described using the classical electrodynamics, and the light scattering causing the phase breaking is treated as a multichannel scattering problem employing a model introduced by Büttiker to describe dephasing in electron transport. We show that the dephasing causes a gradual erosion of the cavity photon mode and thus impedes the formation of molecular polaritons. Consequently, the polaritons signatures in the transmission and absorption optical spectra of the microcavity are fading away as the scattering intensifies.
Read the full abstract

Representation Theory

Graded Kronecker Modules of Constant Rank

Let $K(n)$ be the generalized Kronecker quiver, and let $(T(n),\Omega)$ be the universal covering of $K(n)$, where $n\geq 3$ and $\Omega$ is a bipartite orientation. We introduce three different types of representations for quiver $(T(n),\Omega)$ by $K(n)$: maps are injective in a representation maps are all surjective in a representation, and representations are of constant rank. We use $\Inj, \Sur$ and $\GR$ to denote the corresponding subcategories containing these representations, respectively, and we show that \begin{center} $\GR= \Inj\cup\Sur$. \end{center}
Read the full abstract

Machine Learning

CoArena: Evaluating Computer-Use and Multi-Agent Systems in Real Time

Static benchmarks for computer-use agents fix a task set at release and score every system against it once. That makes them reproducible, and it lets them drift from what they should measure: a fixed task set ages, leaks into training corpora, and cannot follow how people actually use agents from week to week. CoArena measures use directly. Real users submit tasks; two systems, each a single model or a multi-agent pipeline behind the same tool interface, execute the same task concurrently in identical sandboxed desktops; users judge the two outcomes without knowing which system produced them; and a public leaderboard is refit from those judgments. The central contribution is a formal account of what makes such an evaluation real-time. We define real-time as five measurable properties, each with an equation and a worked example: continuous task arrival, live concurrent execution, online rating updates, freshness with contamination resistance, and bounded feedback latency from a failed run to a reusable training environment. The rating methodology follows in full: the Bradley-Terry pairwise model, its likelihood with weighted observations and ties, the penalized maximum-likelihood estimator, and the streaming update applied when a single vote arrives (a stochastic-gradient step on the same likelihood, recovering Elo). It gives confidence intervals from the observed information and a cluster-robust sandwich, rank bands from a parametric bootstrap, the rule by which a new system enters the board, and the convergence rate of the estimate. Vote quality is treated with inter-judge agreement statistics, redundant judging, and explicit handling of ties and abstentions. A five-system example with 211 votes is carried from the vote matrix to ratings, intervals, and rank bands. Every number is derived from stated inputs or labeled illustrative; none is a measurement of a deployed system.
Read the full abstract

Representation Theory

Elementary modules of the generalized Kronecker quivers K(n)

Let $k$ be an algebraically closed field. The generalized or $n$-Kronecker quiver $K(n)$ is the quiver with two vertices, called a source and a sink, and $n$ arrows from source to sink. We use $\modd\cK_n$ to denote the category of the finite-dimensional modules of the path algebra $\cK_n=kK(n)$. There exist a function called dimension vector $\dimu: \modd \cK_n\rightarrow \mathbb{Z}^2; M \mapsto (\dim_k M_1, \dim_k M_2 )$, and two functors $\sigma: \mathbb{Z}^2 \rightarrow \mathbb{Z}^2; (x,y)\mapsto (nx-y,x)$, and $\delta: \mathbb{Z}^2\rightarrow \mathbb{Z}^2;(x,y)\mapsto (y,x)$. Then the set $\mathbf{F}=\{(x,y)\mid \frac{2}{n}x\leq y\leq x\}$ is the fundamental domain of the dimension vectors of regular modules under the action of $\delta$ and $\sigma$ in $\modd\cK_n$. We call a regular module $M\in \modd\cK_n$ elementary if there does not exist a short exact sequence $(0)\rightarrow L\rightarrow M\rightarrow N\rightarrow (0)$ with $L,N$ being non-zero regular. In this note, we focus on the set $\mathbf{F}$, and when $x<n$, we establish a one-to-one correspondence between elementary modules and spaces of fixed rank.
Read the full abstract

Distributed Computing

OpWeave: Flexible Operator Disaggregation for Heterogeneous LLM Serving

LLM serving systems increasingly disaggregate inference into finer-grained stages, with recent approaches separating attention from FFN or MoE execution during decode. This operator-level disaggregated serving (ODS) can improve hardware matching and enable independent scaling, particularly across heterogeneous devices. However, existing systems fix operator boundaries and lack a unified characterization of when disaggregation reduces serving cost. We present OpWeave, an end-to-end framework for heterogeneous ODS. OpWeave provides an analytical cost model that bounds the gains of homogeneous and heterogeneous ODS over colocated serving. It jointly optimizes operator partitioning and deployment configuration through a regularity-aware planner that keeps the search tractable even for hybrid-attention models. A vLLM-based runtime executes the synthesized plans with flexible operator stages across heterogeneous device groups. In our evaluation, OpWeave reduces serving cost by up to $1.78\times$ on homogeneous and $1.89\times$ on heterogeneous GPU clusters relative to the best feasible baseline, while meeting latency SLOs.
Read the full abstract

All authors Zikun Li, Yixuan Mei, Shiqi Pan, Zixuan Chen, Xiaowen Zhang, Mengdi Wu, Shuhuai Lin, Yutong Yang, Zhihao Zhang, Xupeng Miao, Rashmi Vinayak, Zhihao Jia

Human-Computer Interaction

Assessing the Applicability of Existing Design Recommendations to AI Companion Design: A Multi-Method Study

With the rapid proliferation of large language model (LLM)-based systems, AI companions have emerged as conversational agents designed to cultivate emotional connection rather than primarily to support humans in instrumental tasks. Because engagement with AI companions involves relational, emotional, and potentially long-term interactions, their design is consequential. Prior work has offered guidance for designing trustworthy and relational AI systems and has begun to examine design for AI companionship. However, while such work provides insights into possible design solutions, less is known about what makes AI companion design difficult as a design problem. To examine this challenge, we assessed the applicability of existing design recommendations from adjacent domains in the context of AI companion design. Our multi-method investigation unfolded across four phases: literature review, practitioner co-analysis, internal heuristic evaluation, and external expert assessment. Throughout this process, we synthesized nine design principle areas that surfaced tensions in the applicability of existing recommendations to AI companion design. Our findings show that ethical and UX-oriented considerations are deeply intertwined and often require context-sensitive application. We document a systematic, multi-method problem analysis that uses these principle areas as an analytic artifact to examine why existing recommendations cannot be directly transferred to AI companion contexts.
Read the full abstract

All authors Soobin Cho, Deveshi Modi, Divya Mavinkurve, Jieqiong Ding, Mark Zachry

Optimization and Control

Optimal Deterministic First-Order Oracle Complexity for Nonconvex-Concave Minimax Optimization

We study the deterministic first-order oracle complexity of smooth nonconvex-concave minimax optimization over a bounded convex dual domain under two notions of stationarity: optimization stationarity (OS) and game stationarity (GS). OS concerns stationarity of the value function, viewed as a structured nonsmooth objective obtained by maximizing over the dual variable. GS concerns the joint first-order optimality conditions for both primal minimization and dual maximization, each with the other variable held fixed. Existing upper bounds on the oracle complexity of first-order methods for finding $\epsilon$-OS and $\epsilon$-GS points have different dependence on $\epsilon$, but whether this difference reflects an intrinsic separation in computational hardness remains unresolved. We resolve this question by establishing worst-case lower bounds for arbitrary deterministic first-order methods: $\Omega(\epsilon^{-3})$ oracle queries for finding an $\epsilon$-OS point and $\Omega(\epsilon^{-5/2})$ for finding an $\epsilon$-GS point, with the remaining problem parameters fixed. These bounds match existing upper bounds up to logarithmic factors. To close the remaining logarithmic gap, we introduce Tracked-FOAM, a new first-order algorithm whose oracle complexity matches our lower bounds for both OS and GS. Together, these results establish an intrinsic separation between OS and GS in the optimal accuracy dependence of deterministic first-order oracle complexity.
Read the full abstract

Machine Learning

Graph-Transformer Fraud Detection with Self-Supervised Pretraining and Conformal Risk Control

Financial fraud in corporate transaction networks has grown more coordinated and harder to detect with rule-based engines and with classical learning models that treat each transaction in isolation. This paper presents GTFD, a graph-transformer fraud detector that fuses structural and temporal evidence from a corporation's payment graph. GTFD encodes the graph with a multi-head graph attention network, encodes ordered transaction sequences with a gated transformer, and combines both views through a cross-modal gating layer. A conformal risk-control head converts the fused representation into threshold-free anomaly scores with finite-sample coverage guarantees, and the network is trained with self-supervised link-mask pretraining plus adversarial augmentation so it remains stable under scarce labels and under adversarial perturbation. On a corporate transaction benchmark enriched with coordinated fraud rings, GTFD reaches an AUROC of 0.990, an F1-score of 96.1% (precision 96.3%, recall 95.9%), and an accuracy of 98.4%. It reduces the false-positive rate by about 29% relative to the strongest baseline while raising coordinated fraud-ring recall from 85.1% to 96.5%. Ablations attribute roughly 2.0 AUROC points to self-supervised pretraining and 1.9 AUROC points to the conformal head, and adversarial stress tests show GTFD retains 89.2% accuracy at perturbation magnitude 0.20 where the next-best model falls to 76.4%.
Read the full abstract

Optimization and Control

Lower Bounds for Nonconvex-Concave Minimax Optimization

We study lower bounds on the first-order oracle complexity of smooth nonconvex-concave minimax optimization. We consider objectives $f$ that are jointly $L$-smooth in the primal and dual variables $(x,y)$, concave in $y$, and whose primal value function $\Phi(x) := \max_{y\in\mathcal Y} f(x,y)$ satisfies the initial-gap condition $\Phi(0)-\inf_{x\in\mathcal X}\Phi(x)\le \Delta_\Phi$, with a bounded dual domain satisfying $\operatorname{diam}(\mathcal Y)\le D_{\mathcal Y}$. We measure stationarity by the norm of the gradient of the Moreau envelope of $\Phi+\iota_{\mathcal X}$ with parameter $1/(2L)$. We prove that any deterministic zero-respecting first-order algorithm requires $\Omega\left(L^2D_{\mathcal Y}\Delta_\Phi\epsilon^{-3}\right)$ oracle evaluations to find an $\epsilon$-stationary point. Under an unbiased stochastic first-order oracle with bounded variance, any stochastic zero-respecting algorithm requires $\Omega\left(L^3D_{\mathcal Y}^2\Delta_\Phi\epsilon^{-6}\right)$ oracle evaluations. The same lower bounds hold when $\Delta_\Phi$ is replaced by the initial primal-dual gap $\mathcal G_0$. These deterministic and stochastic lower bounds match the corresponding upper bounds of [14] and [29], respectively, up to a logarithmic factor in the deterministic setting.
Read the full abstract

Signal Processing

Gaussian-trigonometric functional link artificial neural network: design and analysis

This paper proposes a Gaussian function-based trigonometric functional link artificial neural network (GTFLN) filter for linear-in-the-parameters nonlinear filtering. Compared with the adaptive exponential TFLN (AETFLN) filter, the GTFLN filter provides smooth and localized basis functions with reduced computational complexity, where modeling advantages are theoretically established through the smoothness, reproducing kernel Hilbert space, approximation error, and operator theory properties. To maximize the modeling performance, an optimized scaling parameter for the GTFLN filter is derived, yielding the optimized GTFLN (OGTFLN) filter. Least mean square (LMS) adaptation is applied to the GTFLN and OGTFLN filters for nonlinear system identification, resulting in the GTFLMS and OGTFLMS algorithms, respectively. Moreover, the theoretical steady-state excess mean-square error of the GTFLN filter is analyzed. Simulations validate the effectiveness of the theoretical analysis and demonstrate the improved performance of the GTFLN and OGTFLN filters over the linear-in-the-parameters benchmarks in nonlinear system identification and nonlinear acoustic echo cancellation. Based on the GTFLN filter, the filtered-g LMS (FgLMS) algorithm is proposed for nonlinear active noise control. Simulations demonstrate improved stability and noise reduction performance compared to the benchmarks.
Read the full abstract

All authors Jie Wang, Lu Lu, Yi Yu, Xiaodong Li, Chengshi Zheng, Rodrigo C. de Lamare

Audio and Speech Processing

Modeling, Scaling, and Decoding: Optimizing Controllable Speech Generation with Nonverbal Vocalizations

Controllable synthesis of nonverbal vocalizations (NVVs) is es- sential for natural and expressive speech, but remains challeng- ing due to their acoustic diversity and imbalanced distribution in existing corpora. To address these challenges, we develop an NVV-aware DiTAR system that models continuous speech latents, encodes the 16 target NVV categories as dedicated to- kens, and adapts stop prediction to distinguish mid-utterance vocalizations from utterance boundaries. Training begins with large-scale bilingual pre-training on diverse NVV speech, fol- lowed by continued supervised fine-tuning on a corpus en- hanced through targeted synthetic augmentation and frequency- aware rebalancing. At inference time, we select the acoustic prompt, tune the LM-guidance and noise-injection scales, and apply Best-of-N sampling with multi-metric selection to re- duce generation failures. The final system achieves an official weighted bilingual score of 62.786, ranking first in Mandarin, second in English, and first overall among participating systems in Track 2 of the ISCSLP 2026 NVVSpeech Challenge. Ab- lation studies show that targeted augmentation benefits under- represented NVV categories the most, while robust candidate selection requires balancing NVV correctness, lexical fidelity, and perceptual quality.
Read the full abstract

All authors Ziyu Zhang, Yun Chen, Taihui Wang, Hanzhao Li, Qicong Xie, Rilin Chen, Zhixian Zhao, Lei Xie

Quantum Physics

An All-van-der-Waals Qubit

Advances in solid-state physics, materials science, and device engineering have accelerated the development of superconducting qubits. Among emerging platforms, van der Waals (vdW) materials and their heterostructures are potentially attractive building blocks for quantum devices, yet their realization in qubit architectures remains largely underexplored. Here we report an all-vdW superconducting qubit based on a NbSe$_2$-hBN-NbSe$_2$ junction, in which a thin hBN layer simultaneously provides Josephson coupling and capacitive shunting between two NbSe$_2$ islands, forming a "merged-element" transmon. Temporal characterization using circuit quantum electrodynamics (cQED) techniques yields an average energy-relaxation time $T_{1,\mathrm{avg}} = 55 ~\mu s$, Hahn-echo coherence time $T_{2\mathrm{E},\mathrm{avg}} = 21 ~\mu s$, and Ramsey coherence time $T_{2\mathrm{R},\mathrm{avg}} = 1.9~\mu s$. The relatively low Ramsey time is primarily attributable to an enhanced sensitivity to charge noise consistent with the realized device parameters and not a fundamental limitation. These results show that lumped-element superconducting qubits based on vdW heterostructures can achieve coherence times comparable to those of conventional Al-AlO$_\mathrm{x}$-Al qubits, while offering a reduced device footprint and suppressed stray capacitive coupling.
Read the full abstract

All authors Sein Park, Sameia Zaman, Junghyun Kim, Junyoung An, Daniel Rodan-Legrain, Hung-Yu Tsao, Chia-Chin Tsai, Aranya Goswami, Réouven Assouly, William P. Banner, Gabriel D. Cutter, Kenji Watanabe, Takashi Taniguchi, Terry P. Orlando, Gil-Ho Lee, Kyle Serniak, Max Hays, Jeffrey A. Grover, Philip Kim, Pablo Jarillo-Herrero, Joel Î-j. Wang, William D. Oliver

Computational Engineering

Topology optimisation and simulated state reconstruction for self-sensing piezoresistive grippers

We consider the computational design of a self-sensing robotic gripper that has the potential to reduce reliance on traditional external sensors. We computationally investigate the potential of a novel sensing method that exploits the coupled electromechanical properties of piezoresistive materials to enable mechanical perception using sparse electrical measurements. In this theoretical proof-of-concept, we show that under the idealised assumptions of linear constitutive relationships and quasi-static gripper operation, it is possible to reconstruct simple contact shapes and gripper displacement fields using a small number of simulated electrical measurements, suggesting the potential of the proposed sensing technique. We use topology optimisation to promote the quality of displacement state recovery by adding an electrical performance objective to the mechanical compliance requirement of the gripper. Using the proposed reconstruction technique, we showcase the improved reconstruction ability of an electrically optimised gripper design through an increased robustness to noisy synthetic measurements compared to a gripper designed with only a mechanical objective in the topology optimisation problem. In the future, detailed inclusion of non-linear effects and experimental realisation of this theoretical proof-of-concept may help remove the constraints of traditional sensors for robotic handling, potentially facilitating sensing in settings where current sensing systems are impractical.
Read the full abstract

All authors Connor N Mallon, Zachary J Wegert, Anthony P Roberts, Joshua Pinskier, Harry Bowman, Vivien J Challis

Strongly Correlated Electrons

Vestigial altermagnetism

Interaction effects and fluctuations govern the phase diagram of systems with multiple competing phases. In particular, as individual phases are destroyed by increasing temperature, fluctuations can still produce a composite `vestigial' order given by a product of individual order parameters. Here, we describe vestigial Altermagnetism (AM) from the thermal melting of antiferromagnetic (AFM) and orbital orders (OO) on the square lattice. The coexistence of AFM and OO can give rise to conventional AM with ordering wave vector $\mathbf{Q}=(\pi,\pi)$. We show that a distinct `vestigial AM' phase with $\mathbf{Q}=0$ can arise from thermal melting of AFM and OO, and investigate its generic phase diagram. We find that the transition from the high-temperature disordered phase into the vestigial AM phase can be either first or second order, depending on the AFM and OO bare susceptibilities and their interactions. We then investigate the unique experimental signatures by calculating the electron spectral function for a minimal two-orbital model. Despite the fact that the vestigial AM order parameter does not couple to electrons directly, fluctuations dominate the self-energy corrections giving rise to a spin- and orbital-dependent quasi-particle lifetime with the AM symmetry.
Read the full abstract

Social and Information Networks

Enhancing Human Mobility Prediction with Spatially Aware LLM-based Multi-Agent Systems

Predicting a user's next POI is a task in human mobility modeling, yet LLM-based approaches focus on semantic reasoning from previous mobility records, while neglecting real-world spatial context. However, human mobility is inherently shaped by spatial cognition, including geographic distance and neighborhood context. This issue is further compounded by prior evidence that LLMs often struggle with spatial reasoning tasks, including distance estimation and geographically biased prediction. To address these limitations, we propose our framework, a multi-agent LLM framework that decomposes next-POI prediction into three stages: Firstly, a Pattern Extraction Agent that captures temporal and categorical mobility patterns from trajectory history; Secondly, a Spatial Reasoning Agent that structures candidate activity choices by combining behavioral preferences with real-world spatial constraints, including geographic distance, road network distance, and neighborhood affiliation; and Thirdly, a Decision Synthesis Agent that integrates behavioral patterns and spatial reasoning for final prediction. Experiments on the NYC benchmark dataset with two LLM backbones show improvements over baseline methods, with up to 493% Hit@1 improvement and 37% relative improvement in Hit@5. Ablations show that combining neighborhood affiliation with distance-based features generally outperforms distance-only settings, and that the Spatial Reasoning Agent plays a crucial role in final prediction by integrating behavioral preferences with real-world spatial constraints, especially for smaller models. Overall, the results highlight the importance of spatial reasoning in mobility prediction. Accurate next-POI prediction requires combining behavioral patterns with explicit real-world spatial constraints, and multi-agent decomposition provides an effective structure for organizing these forms of context.
Read the full abstract

Computation and Language

Corpus Characterization and Inverse Constitutional Fine-Tuning for Style-Aware Radiology Reports

Automated radiology report generation has advanced rapidly in diagnostic accuracy, yet generated reports frequently diverge from the stylistic conventions of authentic radiologist writing in structure, diction, and uncertainty language, a gap which has direct implications for clinician trust and user experience. To address this, we characterize stylistic variation across 2,000 reports from the CheXpert Plus dataset using Bio-ClinicalBERT embeddings, UMAP dimensionality reduction, and HDBSCAN clustering, identifying five distinct reporting patterns differing in pathology focus, narrative structure, and lexical preference. Drawing on these findings, we adapt the inverse constitutional AI framework to derive a style-focused constitution from radiologist-written report pairs without requiring a formal preference dataset. This constitution, encoding conventions of tone, diction, uncertainty calibration, and report structure, is incorporated into the supervised fine-tuning of a MedGemma-4B base model on 25,245 CheXpert Plus training pairs. Constitutional fine-tuning produces a substantial increases in text alignment (BLEU-4: 0.006 to 0.308; ROUGE-L: 0.171 to 0.484) relative to the untuned baseline. These gains show a qualitative shift in structural and lexical alignment rather than marginal improvement, as the baseline model produces near-zero scores due to format mismatch. Overall, we establish corpus-level style characterization and constitutional modeling as an effective and data-efficient strategy for producing radiology reports that conform to authentic radiologist writing conventions.
Read the full abstract

All authors Sarah Y. Li, Elijah Renner, Rayan Ansari, Alaa Youssef

Statistics Theory

Exact Likelihood-Coin Poisson Sampling for Bayesian Inverse Problems with Sharp Complexity Bounds

We develop an exact posterior-sampling framework for Bayesian inverse problems when selected bounded forward observables can be accessed through Bernoulli events. A Bernstein–Poisson construction converts these forward coins into scaled Gaussian likelihood coins, and thinning an inflated prior Poisson point process yields posterior atoms that are iid conditional on their number. For independent Gaussian observations, we derive an exact mean-work identity for the implemented early-stopped factory and sharp small-noise complexity laws governed by local prior-predictive mass near the exact-fit set; a factorial-moment construction extends the likelihood factory to correlated Gaussian errors. The posterior algorithm is model-agnostic once Bernoulli access is available. As one continuum realization, we use Feynman–Kac sampling, Poisson killing, and lazy random-series evaluation for a bounded elliptic resolvent problem with a function-valued coefficient. Numerical experiments validate the forward and likelihood coins, support the predicted work regimes, and demonstrate posterior sampling without deterministic spatial discretization or fixed parameter truncation in the target. A matched-accuracy benchmark against finite-difference prior rejection illustrates how deterministic discretization bias changes the posterior accuracy–cost balance.
Read the full abstract

Number Theory

A Central Limit Theorem for Coprime Pair Counts in Randomly Translated Disks

We study the distribution of the number of coprime lattice points in randomly translated disks. We prove that its spatial variance is asymptotic to $\sigma^2r\log r$ and establish a central limit theorem under the corresponding normalization, where $\sigma^2$ is an explicit positive constant. By truncating the numerators of rational frequencies in lowest terms, the proof reduces higher-moment estimates to counting squarefree denominators in zero-sum relations.
Read the full abstract

Strongly Correlated Electrons

Emergence of Subdominant Quadrupolar Order in the 5$d^2$ Multipolar Mott Insulator Ba$_2$CaOsO$_6$

Relativistic $5d$ Mott insulators host a rich variety of exotic hidden multipolar order, whose microscopic nature remains heavily debated. A broken local point-symmetry (BLPS) phase followed by antiferromagnetism was previously observed in the doped compound Ba$_2$Na$_{0.1}$Ca$_{0.9}$OsO$_6$, where mixed Os valence places the system toward the $5d^2$ end of the $5d^1$-$5d^2$ regime. In contrast, pristine $5d^2$ Ba$_2$CaOsO$_6$ has shown a single octupolar transition with average cubic symmetry, raising the question of whether BLPS arises from chemical disorder in the doped systems. Here, we discover the BLPS phase in pristine Ba$_2$CaOsO$_6$ emerging at $T^* \approx 50$ K, precisely coincident with the octupolar transition. Beyond standard ${}^{43}$Ca nuclear magnetic resonance (NMR) spectral analysis, our spin-echo nutation measurements detect a finite electric field gradient, demonstrating spontaneous symmetry breaking despite the cubic average structure. While our NMR data support the previously identified octupolar order, the BLPS phase reveals a subdominant quadrupolar contribution, establishing that the hidden order in Ba$_2$CaOsO$_6$ is intrinsically multicomponent. Such mixed multipolar states lie beyond conventional theoretical descriptions of the $5d^2$ ground state. Our results establish local symmetry breaking as an intrinsic property of the clean $5d^2$ limit and provide a new constraint on microscopic theories of strongly spin-orbit-coupled Mott insulators.
Read the full abstract

All authors Ilija K. Nikolov, Giuseppe Allodi, Ginevra Corsale, Anshu Kataria, Cierra Foster, Patrick Woodward, Samuele Sanna, Vesna F. Mitrović

Applied Physics

Bimorph Lithium Niobate Thickness-Shear Overtone Film Bulk Acoustic Resonator

High quality factor ($Q$) and overtone operation enable narrow-linewidth acoustic devices with multiple discrete frequencies in a single cavity. Maintaining both high $Q$ and sufficient electromechanical coupling at higher mode orders remains challenging. Here, we demonstrate a bimorph periodically poled piezoelectric film (P3F) lithium niobate (LN) platform for high-order thickness-shear (TS) overtone excitation. The device comprises a bonded 80-$\mu$m-thick single-crystal X-cut LN bimorph with opposite polarizations, patterned top and floating bottom electrodes, and a suspended air cavity. The P3F configuration mitigates charge cancellation from the alternating stress distribution of higher-order TS modes, enabling measurable coupling across a broad sequence of overtones. The thick LN acoustic cavity and increasingly confined high-order mode profiles support low-loss operation. Measured TS overtones extend to 1.75 GHz. At room temperature, representative overtones at 0.77 and 0.89 GHz exhibit 3-dB $Q$ values of 11,338 and 11,917, corresponding to $fQ$ products of $8.74\times10^{12}$ and $1.06\times10^{13}$ Hz, respectively. Cooling from 297 to 12 K systematically enhances $Q$, yielding a peak 3-dB $Q$ of 20,507 at 779 MHz and a maximum $fQ$ product of $1.98\times10^{13}$ Hz at 1.379 GHz. These results establish bimorph P3F LN as a promising platform for high-$Q$, frequency-scalable micro-acoustic resonators in the sub-GHz and low-GHz regimes.
Read the full abstract

All authors Ziqian Yao, Ian Anderson, Tzu-Hsuan Hsu, Vakhtang Chulukhadze, Jack Kramer, Ruochen Lu

Algebraic Topology

Uniqueness of differential Poincaré duality models

We prove the remaining even-dimensional case of the Lambrechts–Stanley "uniqueness" conjecture for 1-connected differential Poincaré duality models. Together with our previous odd-dimensional result, this shows that any two such algebras weakly homotopy equivalent as PDGAs admit direct PDGA quasi-isomorphisms into a common connected differential Poincaré duality algebra. We use our previous method of obtaining differential Poincaré duality models as nondegenerate quotients of Hodge extensions, the new ingredient being a Hodge extension in the middle degree.
Read the full abstract

Rings and Algebras

Yang-Baxter permutation group actions on distributive Yang-Baxter algebras

Let $(X,r)$ be a finite square-free distributive set-theoretic solution of the Yang-Baxter equation, and let $\mathcal{A}(X,r)$ be its associated Yang-Baxter algebra. We prove that $\mathcal{A}(X,r)$ is isomorphic to a skew polynomial algebra and explicitly compute its Nakayama automorphism. We study the natural faithful action of the permutation group $\mathcal{G}(X,r)$ on $\mathcal{A}(X,r)$ by graded algebra automorphisms. We characterize when $\mathcal{G}(X,r)$ is a reflection group and, in that case, describe the invariant subalgebra $\mathcal{A}(X,r)^{\mathcal{G}(X,r)}$ together with its Jacobian, reflection arrangement, and discriminant. We further establish a necessary and sufficient condition for the noncommutative Auslander theorem to hold for the permutation group action on a distributive Yang-Baxter algebra. Finally, for a class of non-trivial distributive Yang-Baxter algebras, we obtain a lower bound for the pertinency of the permutation group action.
Read the full abstract

Robotics

Task-Specified Active Metrological Inspection with Measurement-Steered VLA Manipulation and Deterministic Evidence Gating

High-mix low-volume (HMLV) manufacturing requires inspection systems to adapt to changing parts, specifications, and work orders without repeated task-specific programming. Existing inspection automation typically assumes predefined sensing sequences, while general purpose robot agents optimize task completion rather than the completeness and validity of metrological evidence. We formulate task-specified active metrological inspection and propose From Requirements to Admissible Metrological Evidence (FRAME), a hierarchical dual-arm framework that converts an inspection instruction and structured specification into traceable conformance evidence. FRAME coordinates learned manipulation with calibrated laser profilometry: a task manager grounds and schedules requirements, active surface correspondence verifies physical-to-specification localization, and evidence memory tracks measurement provenance, admissibility, and coverage. Learned components may propose inspection targets and physical access actions, but deterministic datum-grounded measurement, admissibility checks, coverage auditing, and conformance evaluation prevent incomplete or unverified evidence from authorizing PASS. A series of physical experiments shows that FRAME achieves higher end-to-end inspection reliability, fewer false accepts, and shorter task completion time.
Read the full abstract

All authors Zhiling Chen, Jingzhan Ge, Ruimin Chen, Matthew P. Castanier, David Gorsich, Farhad Imani

General Relativity and Quantum Cosmology

Simulation of PBH formation in a matter-dominated universe

We investigate primordial black hole (PBH) formation during an early matter-dominated era using fully nonlinear numerical relativity. The initial condition is set by a functional form of the curvature perturbation including ellipticity, which makes the configuration triaxial. Two kinds of matter descriptions are considered: dust fluid and collisionless particles. In the dust fluid description, numerical computation crashes associated with the appearance of a singularity at which the fluid density diverges, unless the singularity is hidden well inside the apparent horizon. We found that, for the dust fluid description, to observe the horizon formation before calculations crash, the initial amplitude must be larger than the previous analytic estimation by a factor of 2. On the other hand, with the particle system description, calculations do not crash, and we may observe black hole formation after subsequent evolution of the system. Then the threshold of black hole formation is significantly smaller than the previous analytic estimation by an order of magnitude.
Read the full abstract

All authors Chul-Moon Yoo, Albert Escrivà, Tomohiro Harada, Kazunori Kohri

Probability

A law of large numbers for Macdonald coherent measures

We resolve a conjecture on the law of large numbers for coherent measures on Young diagrams associated with Macdonald-nonnegative specializations of the algebra of symmetric functions for all $0\le q,t<1$. We prove that row and column lengths divided by the diagram size converge to the corresponding specialization parameters. The proof combines a monomial estimate for skew Macdonald polynomials with exponential tilting and binomial decompositions obtained from the branching rule for the polynomials.
Read the full abstract