Peer Revenue Equity in science

Papers

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

All of arXiv

Sunday 30 August

Machine Learning

BEACON: Behavioral and Semantic Enrichment of AlphaEarth Embeddings through Tri-Modal Contrastive Learning

Geospatial foundation models such as the AlphaEarth Foundation produce compact and globally consistent representations of the Earth's surface that transfer effectively to a wide range of downstream tasks. However, because these models are trained primarily on Earth-observation imagery, their embeddings mainly capture physical and spectral characteristics while encoding human activity and urban function only weakly. To address this limitation, we propose BEACON, a tri-modal contrastive learning framework that aligns three complementary views of urban space: physical representations from AE embeddings, semantic representations from point-of-interest (POI) text, and human behavioral representations from hourly POI visitation, while keeping the deployed representation image-only. Using the Houston Metropolitan Area as a case study area, we evaluated the performance of the BEACON framework on nine downstream tasks, including seven regression and two classification tasks against six baselines (raw coordinates, Space2Vec, SatCLIP, TESSERA, Clay and AlphaEarth), using frozen linear and MLP probes over five seeds. Under a linear probe, BEACON improves relative R^2 over AlphaEarth by up to 43% for obesity prevalence, 34% for poor mental health, and 22% for median household income, while remaining competitive in the prediction of physical and environmental variables. These findings highlight the value of augmenting geospatial foundation models with semantic and behavioral signals, extending their applicability from physical Earth observation to human-centered urban analytics.
Read the full abstract

Fluid Dynamics

Competing instabilities in wide-gap viscoelastic Taylor-Couette flow: Taylor vs. helical vortices

This paper presents a numerical study of the stability of a polymer solution flow between concentric cylinders with a rotating inner cylinder. The case of a small-radius inner cylinder is considered. The fluid motion is described using a specific case of the Kelvin-Voigt model, often referred to as the Oskolkov model. This model is applicable to very dilute polymer solutions, where the retardation time is much smaller than the characteristic time of the problem and elastic forces are much smaller than viscous forces. The stability of the steady-state motion is investigated using a fully nonlinear approach by means of direct numerical simulation of a perturbation introduced as finite-duration white noise. Depending on the Reynolds number, the perturbation either decays or grows. The critical Reynolds numbers obtained for both Newtonian and non-Newtonian fluids are found to be in agreement with the predictions of the linear theory. It is also shown that an increase in the elastic forces makes helical perturbations more dangerous than their axisymmetric counterparts.
Read the full abstract

Optics

Programmable generation of optical skyrmions on a silicon photonic chip

Optical skyrmions, characterized by topologically stable and spatially varying polarization textures, show immense potential for robust optical communications and metrology. However, conventional methods for generating optical Stokes skyrmions rely on bulky free-space optics, strictly constraining both system miniaturization and dynamic reconfigurability. Here, we demonstrate the efficient and programmable generation of optical skyrmions and bimerons using a compact silicon photonic chip. By integrating a programmable Mach–Zehnder interferometer mesh with a multi-dimensional grating emitter, we dynamically control the amplitudes, phases, and polarizations of emitted fundamental and orbital angular momentum modes. This architecture allows on-demand electrical switching among a complete library of optical quasi-particle states, including Néel, Bloch, intermediate, and anti-type skyrmions and bimerons. Experimental full-Stokes polarimetry confirms high-fidelity polarization textures with near-unity skyrmion numbers. Our foundry-compatible platform translates complex topological light generation into simple voltage controls, paving the way for next-generation communication and sensing systems based on optical skyrmions.
Read the full abstract

All authors Mingyuan Zhang, Xiaofu Pan, Wu Zhou, Wenzhang Tian, Zengqi Chen, Yiou Cui, Yijie Shen, Yeyu Tong, Jianqi Hu

Human-Computer Interaction

Towards Effective Generation of Interactive Visualizations with Vibe Coding: An Empirical Study

Constructing interactive visualizations has traditionally required substantial human effort, involving both technical implementation and design decision-making. Recently, vibe coding, a programming paradigm leveraging Large Language Models to generate, interpret, and refactor code from natural language specifications, has emerged as a promising approach to reduce the burden. However, the capabilities and limitations of vibe coding in building interactive visualizations remain unexplored. To address this gap, we conducted a user study with 78 participants that were tasked with constructing interactive visualizations using vibe coding. We further collected users feedback through questionnaires, interviews, and case analyses. Based on this study, we examine (1) the capabilities and (2) user experience of vibe coding in generating interactive visualizations, and (3) the practical human-agent collaboration strategies adopted. Our findings provide the first systematic assessment of vibe coding for interactive visualization construction, revealing both its strengths and limitations, explaining the shift in developer labor and identifying the hybrid collaboration strategies participants adopted. Furthermore, our study offers insights for more intuitive and robust vibe coding practices.
Read the full abstract

All authors Yanshan Zeng, Ruixuan Tu, Zuo Xiang, Lijia Feng, Guozheng Li, Chi Harold Liu

Sound

PhysWave: Physics-Guided Latent Diffusion Models for Controllable Spatial Audio Generation

Text-to-spatial audio generation, such as text-to-First-Order Ambisonics (FOA), provides a convenient way to create spatial audio for billion-dollar gaming and film industries. However, existing text-to-FOA methods are largely data-driven and may produce audio that violates acoustic relations between source direction and distance. They also separate descriptive and parametric control, forcing users to trade usability for precision. In this paper, we present PhysWave, a physics-guided latent diffusion model for controllable text-to-FOA generation. PhysWave unifies natural-language and trajectory control through a shared waypoint-caption representation, and augments diffusion training with two differentiable acoustic priors: spherical-harmonic direction consistency and inverse-square distance consistency. To support dynamic spatial generation, we further construct a 300K-clip FOA dataset with diverse sound categories and source trajectories. Extensive results show that the proposed priors help PhysWave generate spatially consistent FOA audio while maintaining competitive audio quality. Further analyses show that these physics priors improve spatial consistency during training and can also be used as inference-time guidance for training-free spatial refinement.
Read the full abstract

All authors Lingfeng Yao, Chenpei Huang, Xingke Yang, Ziye Geng, Changqing Luo, Hao Wang, Jiang Liu, Miao Pan

Representation Theory

Pseudo-spherical representations and types for covers of split tori

We introduce a notion of pseudo-spherical representations for any Brylinski-Deligne (BD) cover of a split torus over a $p$-adic field. In particular, we do not assume that the cover is tame. We enumerate the pseudo-spherical representations and compute their dimension. When the torus is the maximal split torus of a connected split reductive group $G$, we enumerate a subset of distinguished pseudo-spherical representations, under mild assumptions on the BD datum. We expect these distinguished representations to provide the torus input for a Borel-Casselman Theorem for BD covers of $G$. We have verified this expectation for certain non-tame double covers in a separate work. While investigating the pseudo-spherical representations, we also prove that the restriction of irreducible genuine representations of covering tori to their maximal compact subgroup is multiplicity-free. From this we see that the irreducible genuine representations of this maximal compact subgroup are types in the sense of Bushnell-Kutzko. We conclude by describing the Hecke algebras attached to these types.
Read the full abstract

Robotics

Module Number Adaptive Visual Shape Control for Serial Modular Soft Robots

Image based shape control provides a simple means of controlling the whole body configuration of soft robots. However, existing data driven approaches are typically developed for fixed robot structures and require new control data when the number of modules changes. This paper presents a module number adaptive visual shape control method for serial modular soft pneumatic robots. A controller trained only on single module actuation shape data is reused for robots with one to five modules by decomposing whole body camera images into local module patches. A single common module segmenter localizes individual modules across all tested configurations, while the same local controller is applied to every extracted patch. Geometric data augmentation improves transferability to downstream modules, and a lightweight mask reconstruction network reconstructs a synthetically removed actuator mask channel. Experiments on physical robots demonstrate shape control across varying numbers of modules and under environmental changes and payload loading. The results show that single module control learning enables scalable whole body control without configuration specific control data collection.
Read the full abstract

All authors Kyohei Akamine, Takato Horii, Yusuke Sakaue, Hiroki Ishizuka

High Energy Astrophysical Phenomena

Photon–Dark Matter Elastic Scattering: An Effective-Operator Scan and First Operator-Resolved Sensitivity Estimates from the Galactic Halo

Elastic photon–dark matter scattering attenuates gamma-ray spectra along a line of sight, probing the same operators as dark matter annihilation to photons but at a rate linear, rather than quadratic, in dark matter density. We consider Standard Model gauge-invariant effective operators of mass-dimension 5 to 7, suppressed by a cutoff scale $\Lambda$, coupling scalar, Majorana or Dirac dark matter to the photon. The leading operators with non-vanishing real-photon amplitudes enter at dimension-5 for Dirac dark matter and dimension-7 for Majorana dark matter. In the electroweak-doublet dipole portal, the inelastic splitting invoked to evade direct detection also closes the CMB annihilation bound, leaving attenuation the only one of the three photon-sector probes that survives. Applying this to a pixel-level reanalysis of 17 years of Fermi–LAT Pass~8 data toward the Galactic centre, we derive the first operator-resolved sensitivity estimates for photon–dark matter scattering from the Galactic halo: $\Lambda \simeq 0.32~\mathrm{GeV}$ for the dimension-5 Dirac dipoles, $0.21~\mathrm{GeV}$ for the dimension-6 scalar Rayleigh operator and $0.79$–$1.06~\mathrm{GeV}$ for the dimension-7 Rayleigh family. The reach is weak: it lies below the EFT-validity threshold across the cold dark matter mass range, and is superseded on the dipole plane by CMB and direct-detection constraints. The framework is calibrated against pseudo-experiments, and recomputes the sensitivity for any instrument that provides a per-bin spectrum with uncertainties and a line-of-sight column density.
Read the full abstract

All authors Trinity Rosebud Stenhouse, Asli Acar, Mikhail Bashkanov, Frank F. Deppisch, Chamkaur Ghag, Dan P. Watts

Computational Geometry

Combinatorial maps for hierarchical splines

Hierarchical splines are an important part of multiscale and adaptive isogeometric analysis formulations. The Bézier meshes of these splines are an essential part of their definition and of several important hierarchical spline algorithms, such as adaptive refinement and Bézier extraction. Topological data associated with the Bézier mesh-such as adjacency information-can be used to improve the performance of many of these algorithms as well as downstream applications of the splines, but typical hierarchical spline formulations do not compute the topological data, storing instead just a list of elements. In this work we present algorithms to build a performant topological data structure, namely the combinatorial map, to represent Bézier meshes of hierarchical splines over cubical cell complexes where the refinement levels have conforming Bézier meshes. This includes hierarchical and truncated hierarchical B-splines, as well as subsets of other hierarchical spline formulations. We show the performance characteristics of the construction algorithms of these hierarchical combinatorial maps, as well as an example use case, showing that the topological information can provide up to an order of magnitude reduction in computation time in downstream applications of the splines.
Read the full abstract

Systems and Control

Adaptive Finite-Time Position-Force Control of Teleoperation Systems With Time-Varying Delays Using a Liquid State Machine Uncertainty Estimator

Teleoperation systems are increasingly used in medical, rehabilitation, and remote manipulation applications, where accurate position/force tracking and stable interaction are essential. In such applications, the remote environment may exhibit viscoelasticity, frictional memory, contact transitions, and other dynamic interaction effects, causing the system response to depend not only on the current state but also on its previous evolution. This history dependence, together with communication delays and uncertain nonlinear dynamics, makes accurate uncertainty compensation particularly challenging. Conventional feedforward neural approximators do not inherently retain temporal information, while fully recurrent architectures may introduce additional computational and online training complexity. To address this limitation, this article introduces the first application of a liquid state machine (LSM) to bilateral teleoperation control. A finite-time adaptive controller is developed using a hybrid position/force auxiliary error system with velocity and force filters, while the LSM is employed to estimate uncertain dynamics by exploiting its intrinsic temporal processing and fading-memory capabilities with a simple adaptation mechanism. Closed-loop stability and finite-time convergence are established through a Lyapunov–Krasovskii framework. Simulations in spring–damper and generalized Maxwell viscoelastic environments demonstrate improved position and force tracking and lower mean execution time compared with an RBFNN-based controller.
Read the full abstract

All authors Shayan Akbari Haghighat, Mohammadali Ghaemifar, Armin Attarzadeh, Mohammadreza Piri Sangdeh

Computation and Language

Evaluating LLMs on Conversational Text-to-SQL under Chain Ambiguity and Intent Drift

Recent advances in large language models (LLMs) have established conversational text-to-SQL as a practical interface between users and databases, often involving multiple turns of clarification and revision. However, existing benchmarks primarily evaluate execution accuracy, leaving the unfolding and shifting of user intent across turns largely uncovered. To address this, we introduce TIDE-Bench, a benchmark for conversational text-to-SQL under chain ambiguity and intent drift evaluation, targeting two recurring patterns: chain ambiguity, where an underspecified question triggers layered clarification with conditional dependencies, and intent drift, where the user retracts and replaces a previously committed request element. Built on 514 anchor SQLs from BIRD, TIDE-Bench comprises 1,542 samples and introduces dedicated metrics for chain identification and drift recognition-resolution beyond execution accuracy. Evaluating 12 advanced LLMs reveals a persistent chain identification bottleneck unaffected by clarification frequency, a wide drift recognition-resolution gap, and overlap between failure modes when jointly activated. The corresponding code of TIDE-Bench is released for further research.
Read the full abstract

All authors Yujia Liu, Jiayan Lin, Zijin Hong, Zheng Yuan, Shengyuan Chen, Hao Chen, Qinggang Zhang, Xiao Huang, Feiran Huang

Analysis of PDEs

Time-modulated refocusing in inhomogeneous medium

In an inhomogeneous dissipative medium, different rays experience different round-trip attenuation, so an instantaneous time mirror may retrace rays without producing a balanced point focus. We solve this problem for a microlocally selected back-propagating transverse-magnetic component generated by a short pulsating modulation of a thin shell surrounding an inhomogeneous Debye medium with Ohmic conductivity. An exact decomposition of the Maxwell–Debye field isolates the pulse-generated return, and a semiclassical mode reduction yields a scalar returned operator whose principal symbol separates the temporal pulse response, shell modulation, and accumulated material loss. Our analysis shows that geometric distortion and amplitude attenuation are both involved in wave propagation, but the geometric distortion cancels between the outgoing and returning branches, whereas physical attenuation remains. A normalized directional energy, computed from pre-modulation measurements using the known lossless backward propagator, contains the same leading attenuation factor as the returned symbol. Its reciprocal defines a smooth, nonnegative, physically admissible modulation that equalizes the selected source-side gain up to a relative $O(h)$ error without requiring a complete attenuation model. With such a modulation, the field generated by a point source has a focal-scale leading profile independent of directional attenuation, with a unique maximum at the source and coherent return time. In every fixed focal window, the maximum is displaced by at most $O(h^2)$ in space and time, and the spatial half-amplitude area is $O(h^2)$.
Read the full abstract

High Energy Physics (Theory)

Teleparallel formulation of $E_{8(8)}$ Exceptional Field Theory

The potential of $E_{8(8)} \times \mathbb{R}^+$ exceptional field theory, suited for studying three-dimensional maximally supersymmetric compactifications of higher-dimensional supergravities, is derived in teleparallel formulation. This means that, starting from an ansatz of the most general combination of the quadratic scalar contractions of the intrinsic torsion, which is manifestly invariant under generalised diffeomorphisms, one selects the potential by imposing the invariance under infinitesimal rotation of the maximal compact subgroup of the duality group. The teleparallel formulation makes manifest at geometric level the preeminent role of the intrinsic torsion in studying compactifications of higher-dimensional supergravities, where the intrinsic torsion is identified with the embedding tensor of the resulting gauged supergravity.
Read the full abstract

Computer Vision

NepScript Genesis: Neural Architecture Search for Handwritten Devanagari Digit Synthesis

This paper introduces NepScript Genesis, a Neural Architecture Search (NAS) framework for automated Generative Adversarial Network (GAN) discovery, applied to conditional Devanagari handwritten digit synthesis. We compare five NAS strategies against a carefully constructed Deep Convolutional GAN (DCGAN) baseline (FID=332.28). Architecture selection utilizes a two-stage pipeline guided by a novel domain-aware evaluation metric (Enhanced Score). Results demonstrate that Adaptive Exploration achieves the optimal quality-efficiency trade-off, attaining an FID of 79.12 – a 76.19% improvement over the baseline – and the highest mode coverage among the NAS strategies (Recall=0.531) in under one GPU-hour. Furthermore, we demonstrate that incorporating script-specific structural heuristics into the search phase prevents early-stage mode collapse. In a downstream low-resource evaluation, augmenting 250 real training samples per class with GAN-generated digits from the best NAS model improves CNN classification accuracy from 91.0% to 96.5% (+5.5 percentage points), demonstrating that NAS-optimized synthesis produces digits of sufficient quality to benefit practical recognition pipelines when real data is scarce.
Read the full abstract

Computation and Language

LoGo: Token-Level Dynamic Local-Global Attention

As context lengths scale, attention increasingly becomes a primary computational bottleneck in large language models. Standard Transformers remain powerful but computationally inefficient, as they allocate the same attention budget to every token regardless of its contextual demand. Existing local-global hybrids provide a more efficient alternative by mixing restricted- and full-context attention, but they typically allocate span statically across layers or heads. To address these limitations, we propose LoGo, a token-level dynamic local-global attention mechanism that uses attention span as a direct proxy for attention budget allocation. Each LoGo layer contains coupled local and global branches: all tokens receive efficient local attention over a restricted context window, while a learned gate activates global attention with full-context access only for tokens requiring long-range information. A threshold-based budget controller maintains a target global ratio without auxiliary losses, and a progressive masking schedule stabilizes training before sparse routing takes effect. We further implement query-sparse Triton kernels that convert reduced global-attention computation into practical speedups. Extensive experiments validate LoGo's effectiveness, showing that it preserves the scaling behavior of full-attention Transformers across model sizes. In controlled comparisons, LoGo improves over the full-attention Transformer and matched-budget static local-global hybrids, with clear gains on long-range retrieval. Analysis further shows that LoGo learns interpretable span allocation patterns. These results suggest that learned token-level span allocation is an effective and scalable way to improve the long-context performance-compute trade-off.
Read the full abstract

All authors Yuqi Pan, Zheng Li, Bohao Tang, Zhen Qin, Guoqi Li

Graphics

As-Rigid-As-Possible Deformation of Gaussian Radiance Fields

3D Gaussian Splatting (3DGS) models radiance fields as sparsely distributed 3D Gaussians, providing a compelling solution to novel view synthesis at high resolutions and real-time frame rates. However, deforming objects represented by 3D Gaussians remains a challenging task. Existing methods deform a 3DGS object by editing Gaussians geometrically. These approaches ignore the fact that it is the radiance field that rasterizes and renders the final image. The inconsistency between the deformed 3D Gaussians and the desired radiance field inevitably leads to artifacts in the final results. In this paper, we propose an interactive method for as-rigid-as-possible (ARAP) deformation of the Gaussian radiance fields. Specifically, after performing geometric edits on the Gaussians, we further optimize Gaussians to ensure its rasterization yields a similar result as the deformed radiance field. To facilitate this objective, we design radial features to mathematically describe the radial difference before and after the deformation, which are densely sampled across the radiance field. Additionally, we propose an adaptive anisotropic spatial low-pass filter to prevent aliasing issues during sampling and to preserve the field with the varying non-uniform sampling intervals. Users can interactively employ this tool to achieve large-scale ARAP deformations of the radiance field. Since our method maintains the consistency of the Gaussian radiance field before and after deformation, it avoids artifacts that are common in existing 3DGS deformation frameworks. Meanwhile, our method keeps the high quality and efficiency of 3DGS in rendering.
Read the full abstract

All authors Xinhao Tong, Tianjia Shao, Yanlin Weng, Yin Yang, Kun Zhou

Robotics

AGM: Achievement-Grounded Memory for Closed-Loop Agents with Frozen VLA Policies

Frozen vision-language-action (VLA) policies offer broad manipulation skills but execute open-loop action chunks without tracking task progress, so the agent cannot reliably decide whether to continue, retry, or terminate. External memory is a natural remedy, yet we find it can be harmful when attempted actions are recorded as completed progress: transient execution failures become persistent task-state errors, and such a memory can underperform no progress memory at all. We propose Achievement-Grounded Memory (AGM), a lightweight closed-loop framework for frozen VLA policies. AGM represents a task as a static subgoal sequence with a dynamic progress pointer and advances the pointer on physically verified achievement rather than on attempts. Proprioceptive gripper-load cues decide when to verify; coherent point tracking verifies grasps, and language-conditioned cross-view comparison, read by a single trained 2.43M-parameter verification head, verifies placements. The policy, tracker, and encoder remain frozen, the head is the only trained component, and deployment needs no auxiliary vision-language model. On the RoboMME Counting benchmark, AGM reaches 100.0% on PickXTimes and 84.0% on BinFill, surpassing the strongest memory-augmented baseline by 7.7 points on the four-task average, and the gains carry over to a physical robot, where AGM reaches 100.0% and 82.0%. These results suggest that reliable embodied memory depends more on disciplined state updates than on memory capacity.
Read the full abstract

All authors Hongbo Gao, Zeyu Ni, Xin Wen, Siyu Xu, Ruifeng Li, Rui Xu, Yunke Wang

Quantum Algebra

Bicrossed products of generalized Taft algebras and Radford algebras

Over an algebraically closed field of characteristic $p>0$, we classify all matched pairs between a generalized Taft algebra $T_{N,n,\xi}$ and the Radford algebra $R$. If $p\nmid N$, every matched pair is trivial, and the corresponding bicrossed product is the tensor product Hopf algebra. If $p\mid N$, matched pairs are parametrized by $\beta\in\mathbb F_p$, and the bicrossed products are described by explicit cross relations. Two such bicrossed products are isomorphic as Hopf algebras if and only if their parameters are equal; hence there are exactly $p$ isomorphism classes in this case.
Read the full abstract

Physics and Society

Integrating adaptive human behavior into epidemic models with large language models

Infectious disease transmission is shaped by patterns of human interaction, which adapt as epidemic conditions change. Capturing these context-dependent behaviors remains a fundamental challenge for epidemic models. Here, we recast this challenge by using large language models (LLMs) to represent adaptive human behavior within mechanistic epidemic models. We operationalize this idea through Generative Adaptive Behavioral Layer for Epidemics (GABLE), which adapts LLMs to infer behavioral responses to epidemic and policy conditions and translates them into age-structured contact matrices coupled to a mechanistic epidemic model. Applied to COVID-19 in France, GABLE reproduced responses in population mixing and age-specific contact structures that remained epidemiologically informative. In short-term forecasting, LLM-generated contact matrices outperformed mobility-driven matrices derived from real-world mobility data, with the largest gains at longer horizons. GABLE also extends beyond forecasting to prospective policy evaluation by projecting behavioral and epidemic responses to candidate interventions before implementation. When supplied with subsequently implemented policies, GABLE reproduced epidemic trajectories and generated distinct responses to alternative policy timing and composition. By leveraging LLMs as a flexible behavioral layer, GABLE provides a framework for coupling context-sensitive behavioral generation with epidemic dynamics.
Read the full abstract

All authors Yicheng Mao, Haoyang Li, Rob Deardon, Hongru Du

Analysis of PDEs

Remarks on the two-dimensional Cahn-Hilliard equation forced by divergence of space-time white noise

We consider the two-dimensional Cahn-Hilliard equation forced by divergence of space-time white noise that represents Kawasaki dynamics in conservative form. The standard heuristic argument shows that the solution is a distribution and thus the product within the nonlinear term is ill-defined, making this stochastic partial differential equation challenging. We prove the global-in-time unique solution theory after multiple Da Prato-Debussche tricks where the last equation is solved in a weak formulation. Besides a minimum amount of the standard renormalization procedure to deal with ill-defined products, our proof consists of applications of deterministic analysis tools and taking advantage of the unique structure of the equation, which is crucial. We also prove the local-in-time unique solution theory of a mild solution, demonstrating that our improved estimate gets us arbitrarily close, although not quite, to being able to prove the global-in-time solution theory with the last equation in not only a weak formulation but also a mild formulation. Additionally, our approach can be readily extended to prove the global-in-time existence, although not path-wise uniqueness, of solution to the singular surface quasi-geostrophic equations with the resulting equation after multiple Da Prato-Debussche tricks in weak formulation.
Read the full abstract

Numerical Analysis

Analysis and Approximation of Stochastic Multiscale Subdiffusion Driven by Fractional Gaussian Noise

This paper investigates a stochastic multiscale subdiffusion model driven by fractional Gaussian noise, where the multiscale Abel kernel with variable exponent $\alpha(t)\in(0,1)$ is used to capture multiscale and crossover behavior in anomalous diffusion. The main difficulties of this model lie in the complexity of the multiscale Abel kernel (e.g. non-monotonicity and non-coercivity) and the low regularity caused by the noise. Concerning these issues, we prove the well-posedness and regularity of the mild solutions by means of solution operator approach and a perturbation technique for multiscale Abel kernel. Then both the semidiscrete-in-time and fully-discrete numerical schemes are proposed and analyzed under the low-regularity numerical analysis framework, with proved temporal and spatial convergence rates. Numerical experiments are presented to substantiate the theoretical results.
Read the full abstract

All authors Jincheng Dong, Ning Du, Xu Guo, Mengmeng Liu, Xiangcheng Zheng

Atmospheric and Oceanic Physics

Python-Fortran Hybrid Programming to Fuse AI and Physical Models: Examples of AI-LDA in climate and weather models (Hf2pMDA_v1.0)

AI provides an unprecedented opportunity for advancing physics numerical modeling including data assimilation, which is a highly efficient and critically-important tool for advancing our understanding on Earth system and its applications. At the same time, deep incorporation of AI and physical modeling can make great driving to advance AI by injecting it rich physics from long time physics-based modeling development. However, since such physics models are conventionally coded in Fortran and AI algorithms usually are conveniently designed in Python, difficulties exist to directly incorporate AI algorithms into physics models, vice versa. Here, based on the F2PY protocol, we have developed a procedure that implements an infrastructure which conveniently conducts Hf2pMDA to form a program entity so that AI algorithms and physical models can invoke mutually. As examples, within Hf2pMDA, a climate coupled data assimilation (CDA) system is naturally upgraded to a strongly CDA (SCDA) system, and a 1 km high-resolution weather DA system is conveniently implemented within a multi-layer downscaling model that has multiscale DA in different nesting layers. In the climate SCDA system, a coupled general circulation model (CGCM) and a multiscale filtering algorithm is integrated by a Python main controller (PMC) that calls Fortran CGCM components and Weakly-CDA modules as well as a data-trained SCDA algorithm by latent space autoencoder in Python. In the high-resolution weather DA system, the downscaled model consisting of traditional Fortran DA modules in all mother domains and Python AE DA algorithm in the central child domain is integrated by a PMC that organizes these components. With convenient realization of deep incorporation of any AI algorithm and physics model, the Hf2pMDA has a great potential to make progress on both AI and scientific modeling.
Read the full abstract

All authors Xianrui Zhu, Zikuan Lin, Shaoqing Zhang, Zebin Lu, Songhua Wu, Xiangyun Hou, Zhisheng Xiao, Zhicheng Ren, Jiangyu Li, Jing Xu, Yang Gao, Rixu Hao, Xiaolin Yu, Mingkui Li, Guangliang Liu

Cryptography and Security

Context or Digits? Balancing Memorability and Efficiency in Virtual Reality Authentication

We present Adaptive Direction-Based Authentication (ADBA), a knowledge-based authentication method for Virtual Reality that decouples users' needs temporally by enforcing password creation based on virtual environment context while supporting both context- and digit-based entries during authentication. This design prioritizes memorability for new passwords and offers both efficient and memorable options to support users' evolving needs. We conducted a remote longitudinal study with 66 participants comparing ADBA against 6-digit PINs over 2-3 weeks. The results demonstrated that ADBA achieved superior memorability and lower perceived task load. Interestingly, no participant chose to enter via digits in the study, yet they still perceived ADBA to be highly efficiency despite longer objective entry times. ADBA also provided security benefits through randomly-generated digit representations, though some degree of password homogeneity was observed in specific virtual environments. Our findings suggest that ADBA offers solid advantages to the traditional PINs, and successfully addresses the tradeoffs between efficiency, memorability, and security under the usage scenarios considered in the study.
Read the full abstract

All authors Yuxuan Huang, Qiao Jin, Tongyu Nie, Victoria Interrante, Evan Suma Rosenberg

Computation and Language

The Emergent Symbolic Structure of Artificial Neural Networks

Modern systems in artificial intelligence (AI) somehow excel in domains for which they seem poorly suited. Intelligence has traditionally been modeled as operating over structured combinations of symbols, such as logical formulas. However, the strongest modern AI systems are based on neural networks, which instead represent information in continuous vectors. Vectors seem inadequate for capturing the structure of language, logic, and other cognitive domains, yet neural networks achieve impressive performance in these areas. How do they do it? In this work, we propose a potential answer: Despite appearances, perhaps the internal representations of neural networks implicitly realize symbolic structure. In support of this hypothesis, we show that the vector representations of a variety of neural networks can be closely approximated with symbolic structures: we can replace the network's entire representation-generating process with a closed-form equation instantiating a symbolic structure, and the network's behavior remains largely unchanged. This finding holds for both small-scale neural networks trained to manipulate lists as well as large language models (LLMs) operating in four domains that are central in symbolic traditions: arithmetic, logic, computer code, and language. Further, our symbolic approximation allows us to modify an LLM's behavior in targeted ways via precise interventions on its internal representations, showing that the LLM's behavior is reliant on the symbolic structures we have identified. This work provides a potential way to reconcile longstanding symbolic conceptions of intelligence with the vector-based nature of modern AI.
Read the full abstract

All authors R. Thomas McCoy, Paul Soulos, Tal Linzen, Paul Smolensky

Computation and Language

Argument-Aware Semantic Alignment of Normative Texts: A Toulmin-Based Neuro-Symbolic Approach

Semantic alignment between specialized normative texts is challenging when equivalent requirements use different terms, syntax, and levels of abstraction. Lexical overlap, distributional embeddings, and semantic similarity capture topical relatedness but often miss the argumentative structure by which normative claims are supported, qualified, and justified. This paper asks whether explicit argument structure adds information complementary to neural semantics for aligning requirements. We treat cross-standard control mapping as argument-aware semantic alignment and build a neuro-symbolic pipeline that combines neural text representations with Toulmin features. An LLM explicitation step identifies claims, grounds, warrants, qualifiers, and backing and reconstructs enthymemes. These feed an alignment model via argument-aware similarity and structural features. On a NERC-CIP to NIST-CSF mapping benchmark, argument-derived features improve alignment over a neuro-symbolic semantic baseline. Feature selection shows especially strong signal from warrant-related features, indicating that the link between a claim and its supporting reasoning is not captured by conventional similarity alone. A compact claim–grounds–warrant subset remains competitive with the full Toulmin feature set. The results give preliminary evidence that argument structure is a useful intermediate representation for aligning specialized normative texts. Cybersecurity standards are used as a controlled testbed, not as proof of domain-independent generalization. The argument graphs produced by LLM explicitation may also support later work on retrieval, reasoning, and explanation over normative text.
Read the full abstract

Machine Learning

MedCache: Efficient and Temporally Valid Memory for Longitudinal Clinical Agents

Longitudinal clinical agents must maintain an evolving patient state from evidence distributed across visits, time points, and specialties. However, how agent memory should be designed for this setting remains unclear. We introduce a benchmark of multi-visit, multi-specialty patient records that evaluates long-context evidence retrieval, cross-time evidence aggregation, and cross-specialty clinical reasoning. Using this benchmark, we systematically study four memory design choices: curation, organization, retrieval, and memory-augmented reasoning. We find that temporal validity is more important than simply retaining more history; specialty-factorized memory reduces context but can hide shared evidence; and multiple agents help when specialists must reason together, not merely when evidence comes from multiple memories. Guided by these findings, we propose MedCache, a hybrid framework that constructs temporally valid patient memory, organizes evidence into overlapping specialty views, routes each query to relevant memories, and adaptively invokes one or multiple specialists. Experiments show that MedCache improves reasoning accuracy and memory efficiency over strong single-agent and multi-agent baselines, while generalizing across model backbones and external datasets.
Read the full abstract

All authors Hei Ting Chan, Chenwei Wu, Xueshen Liu, Boyuan Zheng, Liyue Shen, Jiasi Chen, Z. Morley Mao

Number Theory

Lagrange spectrum for Diophantine approximations of complex numbers with real part equal to one half

Motivated by a theorem of A. Schmidt on the part of the complex Lagrange spectrum below $2$, we study the restricted Lagrange spectrum $L_{\frac{1}{2}+i\mathbb{R}}$ arising from the approximation of complex numbers of the form $\frac{1}{2}+i\alpha$, $\alpha\in\mathbb{R}\setminus\mathbb{Q}$ by Gaussian rationals $p/q$ with $p,q\in\mathbb{Z}[i]$, $q\neq 0$. We show that this spectrum admits a description in terms of a dynamical spectrum associated with a real horseshoe. As a consequence, we obtain several fractal properties of $L_{\frac{1}{2}+i\mathbb{R}}$, such as continuity of the dimension function $t\mapsto\dim_H(L_{\frac{1}{2}+i\mathbb{R}}\cap(-\infty,t))$. We also prove that the set of complex numbers $z$ satisfying \begin{equation*} \left\lvert z-\frac{p}{q}\right\rvert\geq\frac{1}{2|q|^2}, \quad\text{for all } p,q\in\mathbb{Z}[i], q\neq 0, \end{equation*} is uncountable. In fact, we show that this inequality holds for every complex number of the form $z=\frac{1}{2}(1+i\theta)$ where $\theta\in\mathbb{R}\setminus\mathbb{Q}$ is a root of one of Schmidt's $C$-minimal forms.
Read the full abstract

All authors Hao Cheng, Harold Erazo, Carlos Gustavo Moreira, Thiago Vasconcelos

Computation and Language

Ontology-Guided Multi-Agent Extraction of Evaluation Objects from Academic Review Texts: Evidence from Chinese Library and Information Science

Academic reviews, scholarly commentaries, and book reviews serve as sources of evaluative statements about theories, methods, literature, institutions, and policies, providing valuable evidence for scholarly evaluation. Existing scientific entity extraction methods mainly target research articles and are less effective for evaluation objects, which are often abstract, context-dependent, and characterized by ambiguous type boundaries. This study proposes an ontology-guided multi-agent framework for evaluation object extraction. The framework combines candidate discovery, ontology-constrained classification, and domain review. Experimental results show that it achieves a Precision of 90.33%, Recall of 84.55%, Entity-level F1 of 87.34%, Strict Typed F1 of 79.78%, and Type Accuracy of 91.35%, substantially outperforming rule-based and zero-shot baselines. Ablation results indicate that the multi-agent workflow improves recall and stability, while ontology-based boundary constraints enhance fine-grained classification and reduce category confusion. The framework supports the structured utilization of evaluative scholarly texts and provides methodological support for evidence-based research evaluation and STI mining.
Read the full abstract

All authors Haolin Chen, Hongyi Dong, Yu Zhu, Yijia Hong, Leiqing Niu, Jiyuan Ye

Combinatorics

Anti-Ramsey Number for Suspension of Edge-Critical Graphs

An edge-colored graph is called a rainbow graph if all its edges have distinct colors. The anti-Ramsey number, denoted by $\ar(n,F),$ for a fixed graph $F$ and a positive integer $n$, is the maximum number of colors used in an edge-coloring of the complete graph $K_n$ that contains no rainbow copy of $F$. Meanwhile, the Turán number, denoted by $\ex(n,F),$ for graph $F$ and $n$, is the maximum number of edges in an $n$-vertex graph that does not contain $F$ as a subgraph. For a vertex $v$ and a multiset $\mathcal{H}$ of graphs, the suspension $\mathcal{H} + v$ of $\mathcal{H}$ is the graph obtained by connecting the vertex $v$ to all vertices of $H$ for each $H \in \mathcal{H}$. Let integers $k\ge 1$ and $r\ge 2$ be fixed, and suppose that $\mathcal{H}_{k+1}=\{H_1, H_2, \ldots, H_{k+1}\}+v$ satisfying $H_1, H_2, \ldots, H_{k+1}$ are pairwise vertex-disjoint edge-critical graphs, and $\chi(H_i)=r$ for $i=1,2,\ldots, k+1$.In this paper, we determine $ \ar(n,\mathcal{H}_{k+1}) $ for $k\ge 1$, $r\ge 2$ and sufficiently large $n$. This result unifies and generalizes a result of Liu et al. (arXiv:2411.08475) concerning the friendship graph, and a result of Lu et al. (arXiv:2507.13165) on the intersecting cliques.
Read the full abstract

Mathematical Physics

Kubo Formulas and Bulk-Edge Correspondence for Curved Boundaries

Strong topological insulators are classified by integer invariants which admit different real-space expressions. Inspired by recent work studying these invariants for spaces with curved boundaries, we revisit the problem of equivalence of the various expressions, including Fredholm index pairings and Kubo formulas involving half-space projections. Using Roe algebras and a variant of KK-theory suitable for non-separable $C^*$-algebras, we prove a general form of bulk-edge correspondence for the index pairing with a general position-space Dirac operator in the presence of arbitrary boundaries. For spaces coarsely equivalent to $\mathbb{R}^d$, we show that up to multiplicity, these pairings are equal to those obtained with the standard dual Dirac operator, with the multiplicity explicitly given as the topological degree of the symbol function. In addition, we prove a generalized Kubo formula which computes the general index pairing by a real-space formula.
Read the full abstract