Soft Condensed Matter
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Thursday 3 September
Jammed suspensions of soft microgel particles may exhibit slippage along smooth boundaries. Owing to their expected sub-micrometric dimensions, direct observations of dynamics within the near-surface layers supposed to be responsible for this slippage have been difficult to achieve. Here, we use total internal reflection fluorescence microscopy (TIRFM) to observe nanoparticle dynamics near glass/microgel-suspension interfaces. Indicating near-wall dynamic heterogeneity, velocity profiles for suspensions are nonlinear. These profiles tend to a constant slippage velocity at submicrometric distances from the wall, consistent with macroscopic wall slip measurements. Furthermore, nanoscale particle altitude distributions are strongly dependent on the slip velocity, revealing a dynamically-mediated and nanoscale particle-organisation effect. The collected observations give support for the existence of near-wall heterogeneity as a dominant mechanism contributing to microgel wall slip. Our work also opens new perspectives for the study of particle dynamics and organisation in complex interfacial environments.
Plasma Physics
Formation and interaction of two-dimensional electron-acoustic solitons and breathers in superthermal plasmas
The nonlinear evolution and mutual interaction of two-dimensional electron acoustic (EA) nonlinear structures in superthermal plasma environment are studied. The plasma model consists of inertial cold electrons, superthermal hot electrons described by kappa ($\kappa$) distribution, and stationary ions providing overall charge neutrality. Using the extended Poincaré-Lighthill-Kuo (PLK) reductive perturbation technique, a pair of two-sided Kadomtsev-Petviashvili (KP) equations governing right- and left-propagating EA solitary waves (EASWs) is derived. Exact analytical solutions of the KP equations, including single soliton, multisoliton, breather, and lump structures, are obtained via the Hirota bilinear method. The effects of key plasma parameters such as hot electron concentration, temperature ratio, and superthermality index on the characteristics of these nonlinear excitations are examined. Particular attention is devoted to the head-on collision dynamics between solitons, breather-soliton, and breather-breather interactions. The results reveal quasi-elastic collisions accompanied by phase shifts, transient amplitude modulation, and localized energy concentration, with clear distinctions between oscillatory and non-oscillatory mode interactions. The present study provides new insights into multidimensional electron acoustic wave (EAW) dynamics and energy redistribution mechanisms in superthermal space plasmas, with direct relevance to planetary magnetospheric environments such as Saturn's ring region.
Read the full abstract
Databases
Tree Databases
We propose a novel database model whose basic structure is a labeled, directed tree with node identities. Intuitively, the root of the tree is seen as an object (or entity), the non-root nodes as attributes of the object and the semantics of each attribute is represented by the unique path leading from the root to the attribute. We define a tree database to be a set of such trees. The trees of the database can be combined to produce new trees using a set of operations on trees that we define in the paper. The query language of our model offers two types of queries, traversal queries and analytic queries. A query (whether traversal or analytic) is always defined over a tree, which is either a tree in the database or a tree derived from other trees using tree operations. The operations on trees and the query language are both defined using a simple functional algebra whose operations are: restriction of a function, composition of functions, pairing of functions and Cartesian product of sets. A distinctive feature of our model is that traversal queries and analytic queries are both defined within the same formal framework; and in fact, traversal queries serve as the building blocks for analytic queries. This is in sharp contrast to the relational model, where analytic queries are defined outside the relational algebra, in the form of SQL Group-by queries. Therefore our model supports data access and data analysis within the same formal framework. We demonstrate the expressive power of our model by showing: (a) how our model can support inheritance in a seamless manner, (b) how one can define consistent relational databases on top of a tree database - with the tree database playing the role of an underlying semantic layer and (c) how a tree database can be used as a user-friendly interface for accessing and analyzing relational data.
Read the full abstract
Astrophysics of Galaxies
Bright star-forming galaxies naturally forming at z>10 in the Shark semi-analytic model
The James Webb Space Telescope (JWST) has unveiled the existence of numerous z>10 bright ultraviolet (UV) galaxies, potentially challenging galaxy formation models in a Lambda Cold Dark Matter (LCDM) universe. Modifications to star formation and stellar feedback models have been suggested to alleviate the tension. However, the fundamental challenge is to design a galaxy formation model that simultaneously reproduces observations from z=0 to those at the highest redshifts. Here, we present predictions from the Shark semi-analytic model of galaxy formation, which is tuned to reproduce the z=0 universe. We show that the same model is capable of reproducing the current UV luminosity function constraints even up to z=17 without the need to invoke variations in the baryon physics model. This model is also capable of reproducing reasonably well the stellar mass function evolution from z=0 to z=10 and the cosmic star formation rate (SFR) density at 0<z<15, again with the same baryon physics and parameters. The key to the success of Shark is the triggering of starbursts by violent disk instabilities included in the model, which leads to an increase in the variance of the UV luminosity with increasing halo mass at z>10. We demonstrate that without this mechanism, galaxies are not bursty enough in the model to reproduce the observations at z>10.
Read the full abstract
Mathematical Physics
Controllability, returning waves and scattering without reverberation in 3D acoustic dynamic system
The dynamic acoustic scattering system is governed by the wave equation $u_{tt}-\Delta u+qu=0$ in $\Bbb R^3,\,\,\,-\infty<t<\infty$, with a compactly supported potential $q$ and infinitely distant sources (controls) $f$, which initiate incoming spherical waves $u=u^f(x,t)$ provided $u^f\big|_{|x|<-t,\,\,\,t<0}=0$. These waves are focused at $x=0$ and fill up the whole space at the moment $t=0$. The system is controllable if the set of waves $u^f(\cdot,0)$ produced by all finite energy controls $f$, covers the space $L_2(\Bbb R^3)$. As we show, if the Hamiltonian $H=-\Delta+q$ has the bound states, then in the space the points $a$ appear such that the system, being refocused at $x=a$, loses controllability. The latter leads to a physical effect: the waves $u^f$ of finite energy appear, which vanish simultaneously in the past and future cones $|x|<\pm\, t$ and leave the region of inhomogeneity of $q$ without reverberation. This effect has some similarities with the wavefront reversal (Time Reversing Mirror), but is more meaningful from a mathematical point of view.
Read the full abstract
Artificial Intelligence
Feature Reconfiguration With Visual Prior for Medical Lesion Segmentation
Lesion segmentation in medical images plays a critical role in clinical diagnosis and treatment planning. Despite significant advances, lesion segmentation remains challenging due to two major factors: (1) complex background interference; (2) diverse lesion morphology. Existing encoder-decoder based methods mainly focus on enhancing feature extraction or redesigning decoding strategies. However, they lack early prior guidance and feature reconfiguration during the encoding stage, limiting their effectiveness in handling these challenges. To address these limitations, we propose FreNet, a feature reconfiguration framework with visual priors, which performs pixel-level reconfiguration before encoding and feature-level reconfiguration during encoding for precise medical lesion segmentation. To suppress background responses, we propose an Implicit Prior Neural Network (IPNN), which models a continuous spatial field and leverages visual prior from SAM to reconfigure input image before encoding stage. To better handle diverse lesion morphology, we design a Dual-domain Feature Reconfiguration (DFR) module to progressively reconfigure backbone features during encoding stage. Within DFR, the Frequency Decoupling Module (FDM) decouples backbone features in frequency domain to enhance foreground-background discriminability, while the Spatial Localization Module (SLM) spatially relocates and improving spatial stability after frequency decoupling. Extensive experiments on 9 medical image segmentation benchmarks across three imaging modalities demonstrate that FreNet significantly outperforms state-of-the-art (SOTA) methods. On the challenging ETIS dataset, our method achieves Dice improvements of 5.0% over SOTA method and 7.2% over SAM.
Read the full abstract
Computer Vision
TruncGradGS: Improved 3D Gaussian Splatting via Truncated Gradient Updates
3D Gaussian Splatting has become a de facto scene representation for novel view synthesis, yet robustly learning 3D Gaussian primitives from visual input remains challenging. Standard optimization relies on gradient-based updates, but a common issue is the gradient vanishing phenomenon: a pixel far from a Gaussian primitive often has diminishing gradient magnitudes to influence primitive attributes, resulting in suboptimal scene reconstruction. In this paper, we propose a method to address gradient vanishing with a piecewise truncated gradient formulation that improves the optimization stability and robustness to initializations. We show that our method consistently improves 3D Gaussian Splatting with random and COLMAP initializations while being generalizable across static and dynamic Gaussian Splatting. As a by-product, we also examine the limitations of current benchmarks for dynamic scenes, and introduce a novel dataset for benchmarking dynamic Gaussian Splatting using synthetic 3D scenes. We demonstrate the effectiveness of our method in both static and dynamic settings for the public benchmarks and our proposed dataset.
Read the full abstract
Machine Learning
Coupled Scaling: A Representational Accessibility Framework for Neural Scaling Laws
We ask when two learning systems trained on the same task under a common resource protocol should share a scaling rate and when their rates should differ. Coupled Scaling answers this through representational accessibility: the task-relevant geometry that a specified architecture-optimization system can reach and how that geometry is acquired as resources grow. In an orthogonal model, unsupported target energy sets the asymptotic floor, while unacquired supported energy sets the finite-budget residual. When acquisition can skip high-value directions, the largest fully acquired prefix no longer determines a unique exponent. For target powers $a_j \asymp j^{-b}$, $b>1$, and prefix log-growth rate $0<\rho\leq1$, the sharp attainable interval is $[\rho(b-1),\min\{b-1,\rho b\}]$, and every rate in this interval is realized by a fixed acquisition order on a common support. A rank-window condition identifies when the product law $\alpha=\rho(b-1)$ applies, while a fixed-kernel specialization expresses the same task-system dependence through the task-weighted spectral tail. We then audit released capability trajectories to diagnose measurement and comparison effects and propose a staged test in which independently measured geometry predicts held-out loss, scaling rates, and cross-task reversals.
Read the full abstract
Probability
Airy limit for the Jack process and topological expansion
For every $\beta>0$, we establish multi-time soft-edge moment convergence for a class of Jack processes, including the Jack–Plancherel process, a discrete $\beta$-analogue of Dyson Brownian motion. The limiting moments admit an absolutely convergent expansion in terms of nonnegative Brownian bridges decorated by pairs of equal and opposite jumps. At equal times, we identify this limit with the joint Laplace statistics of the Airy$_\beta$ process for $\beta\geq1$. The same Brownian expansion yields an asymptotic $\beta$-topological expansion of the marginal $b$-conjecture and, at $\beta=2$, an explicit nonnegative formula for the Witten–Kontsevich intersection numbers.
Read the full abstract
Optimization and Control
Beyond Higher-Pulse Rectification: Operational Harmonic Coordination in Renewable P2H Systems
Thyristor rectifiers (TRs) are cost-effective electrolysis power supplies for renewable power-to-hydrogen (ReP2H) systems, but their harmonics may violate grid-code limits. In contrast to conventional solutions that rely on higher-pulse (such as 24-pulse) rectifiers, this paper proposes an operational harmonic coordination scheme that enables low-cost 12-pulse TRs to meet harmonic requirements through coordinated operation. First, a harmonic model quantifies the effects of rectifier transformer (RCT) tap positions and electrolytic currents, enabling harmonic cancellation among multiple electrolyzers (ELZs). A two-layer framework then coordinates hydrogen production and harmonic mitigation. Hourly scheduling determines ELZ commitment within the harmonic feasible region under renewable uncertainty using stochastic programming and a modified progressive hedging algorithm, while minute-level dispatch tracks renewable power and mitigates harmonics. A decomposition algorithm separates production dispatch from harmonic mitigation to improve computational efficiency. Case studies based on real-life projects show that the proposed method increases profit by 31% over current-only regulation. Annual simulations further show that coordinated 12-pulse TRs can replace 24-pulse rectifiers for harmonic compliance by exchanging additional RCT tap actions for lower transformer investment, reducing rectification-stage cost by 37.5%.
Read the full abstract
High Energy Physics (Theory)
Topics in Kadyshevsky Field Theory
In these notes we deal with several field theoretical topics in the framework of the Quantum Field Theory as developed by Kadyshevsky. The main motivation for studying the Kadyshevsky formalism is that in the kadyshevsky graphs, in contrast with the Feynman graphs, the particles remain on the mass shell. This facilitates the use of phenomenological form factors, e.g. Gaussian ones. In the first part we construct the second quantised quasi-particle formalism, which is employed to develop the functional integral formalism. In the latter we develop the path-integral, the Schwinger-Symanzik equations, generalised Wightman functions, and the Kadyshevsky reduction formulas. In the second part we cover the topics: (i) the relation between the Feynman and Kadyshevsky perturbation theory, and (ii) the Gross-Jackiw method in the Kadyshevsky formalism. The latter method is applied to the kadyshevsky formalism for interaction Lagrangians with derivatives, in particular for pion-nucleon interactions: (i) pseudo-vector $NN\pi$-, (ii) vector $NN\rho$-, and (iii) gauge-invariant $\Delta_{33}N\pi$ coupling.
Read the full abstract
Databases
KnowFeat: Knowledge-Guided Feature Engineering via LLM Agents
Automated feature engineering with large language models (LLMs) can produce semantically meaningful features for tabular data, yet existing methods lack structured domain knowledge, rigorous verification, and explainable provenance. We propose KnowFeat, a knowledge-guided feature engineering framework that organizes domain knowledge into five types – schema metadata, regulatory indicators, detection rules, expert opinions, and court document evidence – and injects them as structured context into an LLM agent. A three-stage verification pipeline filters candidates through code execution, statistical quality checks, and model effectiveness evaluation. Every accepted feature carries a provenance record tracing its design to specific knowledge assets. Under a strict held-out protocol that eliminates feature-selection leakage, KnowFeat ranks first (avg. rank 2.3) across twelve public benchmarks among seven methods (one-sided Wilcoxon p=0.017), with a peak gain of +11.6 pp AUC on a telecom churn dataset. On a real-world Bitcoin anti-money laundering (AML) dataset (Elliptic) and a synthetic digital currency AML benchmark (SimECNY), KnowFeat maintains competitive detection performance with full provenance traceability.
Read the full abstract
Machine Learning
LeanGRPO: Eliminating Redundant Recomputation in Diffusion RL
Diffusion reinforcement learning (RL) has recently achieved significant success in post-training image and video generative models. However, most diffusion RL methods, including DanceGRPO and FlowGRPO, recompute selected timesteps with gradient tracking after rollout. Under on-policy training with the same backend for rollout and update, this recomputation is mathematically redundant. Intuitively, the rollout and policy update steps can reuse the same feed-forward backbone to avoid redundant computation, but doing so can incur a large memory overhead during rollout. To address the issue, we present LeanGRPO by restructuring the data-parallel layout and introducing two recompute-free training schedules for trajectory-logprob diffusion RL: (1) LeanGRPO-Retain enables gradient tracking during rollout and directly reuses the resulting computation graphs and saved activations for backward during update, requiring no recomputation; and (2) LeanGRPO-Reweight also enables gradients during rollout, but immediately backpropagates each selected step using a provisional advantage and delays gradient synchronization, then corrects the provisional gradients with the true advantage after the trajectory is completed. These schedules target different model scales and input sizes. Across FlowGRPO/DanceGRPO with FLUX.1-dev and Wan, LeanGRPO achieves up to 1.83x end-to-end speedup while preserving the original optimization objective.
Read the full abstract
Artificial Intelligence
NeoRed: A Knowledge-Logic-Alignment Multimodal Large Language Model for Neonatal Respiratory Disease Diagnosis
Neonatal respiratory diseases are a major cause of neonatal morbidity and mortality, posing substantial challenges in clinical practice. Despite recent advances, existing Multimodal Large Language Models (MLLMs) face two key limitations in neonatal diagnosis: (1) domain gap arising from predominantly adult training data; (2) insufficient integration of multidimensional clinical context for accurate diagnosis. To address these challenges, we collect two real-world clinical datasets (NeoCXR and NeoCXR-EV) and propose NeoRed, to the best of our knowledge, the first MLLM tailored for neonatal respiratory disease, filling the gap in neonatal diagnostic reports generation. To enhance joint diagnosis from heterogeneous clinical context and chest X-rays, we design a novel Knowledge-Logic-Alignment (KLA) framework which constrains model behavior from three perspectives: 1) Knowledge Prior Injection (KPI) incorporates neonatologist-inspired diagnostic priors into multimodal representations, guiding disease-specific attention across modalities; 2) Diagnostic Logic Constraint (DLC) aligns the semantics of generated reports with multimodal diagnostic logic; and 3) Visual Semantic Alignment (VSA) establishes semantic correspondence between visual features and imaging conclusions. Extensive experiments demonstrate that NeoRed enables accurate neonatal diagnostic reports generation, achieving ROUGE-L of 53.29% and Clinical Efficacy F1 score of 65.19% on NeoCXR, outperforming existing MLLMs. NeoRed also preserves competitive report generation performance on adult benchmarks (MIMIC-CXR and IU-Xray). Datasets will be available upon application.
Read the full abstract
Artificial Intelligence
CulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning
Multimodal language models achieve near-ceiling scores on food recognition benchmarks, yet it remains unclear whether this success reflects genuine cultural understanding or mere visual matching. To probe this distinction, we introduce CulturalMenuBench, a benchmark of 4,870 items in 10 languages across 18 regions; its 10 tasks pair final-dish and step-by-step cooking images with ingredients, procedural text, and regional labels, spanning basic recognition to process-grounded cultural attribution. Evaluating 12 models exposes a substantial knowledge-application gap: models exceeding 94% on standard multiple-choice tasks drop to at most 56% when attributing dishes to Chinese regional cuisines, despite an identical four-way format. Diagnostic analyses explain why: error patterns are consistent with random guessing, accuracy tracks visual distinctiveness rather than cultural structure, and models classify cuisines more accurately from dish names alone than from images (+7-18 points). The knowledge is thus present but cannot be activated through visual input. An ablation confirms these tasks genuinely require procedural evidence: removing sequential cooking images selectively degrades process-grounded tasks while others remain stable. Overall, CulturalMenuBench shows that near-perfect recognition can conceal an inability to apply cultural knowledge, motivating training that explicitly connects perception, procedure, and cultural context. Code and data are publicly available.
Read the full abstract
High Energy Physics (Theory)
Center-twisted Gribov spectra and the finite-volume Gaussian response in the refined Gribov–Zwanziger framework
We develop a continuum framework for comparing the Gribov–Zwanziger and center-vortex descriptions of confinement through the gauge-invariant twisted partition function of the electric $\mathbb{Z}_N^{[1]}$ 1-form symmetry. A background 2-form field $B$, equivalently an 't~Hooft twist on a torus, labels a global sector and is not itself a dynamical center vortex. For a minimal irreducible twist on $T^4$, we derive the complete adjoint momentum lattice of $SU(N)$. The twisted spectrum is exactly the scalar spectrum on an enlarged torus with periods $(L_1,L_2,NL_3,NL_4)$ with the ordinary-torus sublattice removed. This yields a finite Faddeev–Popov gap at the flat representative and reduces twisted-minus-untwisted spectral traces to ordinary torus traces. For the refined Gribov–Zwanziger (RGZ) kernel, Poisson resummation gives an exact finite-volume Bessel-function winding sum with a universal center-twist projector. We evaluate the Gaussian one-loop integral at fixed RGZ parameters up to the finite-dimensional global zero-mode/stabilizer normalization. The Zwanziger determinants cancel, while the gauge-fixing/ghost sector leaves a universal massless primed determinant on $T^4$; the untwisted sector also contains constant gluon modes. The only normalization not fixed by the local quadratic Hessian is the relative zero-mode/stabilizer measure of the reducible untwisted and irreducibly twisted flat connections. We also derive closed finite-volume sources for the RGZ stationary equations. The massive response is exponentially suppressed at large volume, whereas the massless factor depends on the global zero-mode normalization. Thus the Gaussian calculation does not by itself establish the strong center-vortex-condensation criterion for $Z[B]/Z[0]$, but it isolates the remaining global normalization problem and separates a global twist from an actual dynamical vortex.
Read the full abstract
Representation Theory
Exceptional model structures and the induced extriangulated categories
A Hovey triple $(\mathcal{C}, \mathcal{F}, \mathcal{W})$ is exceptional, if $(\mathcal{C} \cap \mathcal{F}, \ \mathcal{C} \cap \mathcal{F} \cap \mathcal{W})$ is not a Frobenius pair. One has a disjoint union $\{\text{Hovey triple}\} = \{\text{Hereditary Hovey triple}\} \ \dot\bigcup \ \{\text{Exceptional Hovey triple}\}$ \ $\dot\bigcup \ \{\text{non-hereditary and non-exceptional Hovey triple}\}.$ Exceptional Hovey triples appear widely. For a selfinjective Nakayama algebra $A= kC_n/J^t$, $A\mbox{-}{\rm mod}$ admits exceptional Hovey triples if and only if $\gcd (n, t) \ge 2$ and $t \geq 3$. Their homotopy categories reveal new phenomena. Nakaoka-Palu's Theorem implies that there is a triangulation on $\frac{\mathcal C\cap\mathcal F}{\mathcal C\cap\mathcal F\cap\mathcal W}.$ It is proved that a Hovey triple in a weakly idempotent complete extriangulated category $(\mathcal A, \mathbb E, \mathfrak s)$ is exceptional if and only if the induced extriangulated structure $\bigl(\frac{\mathcal C\cap\mathcal F}{\mathcal C\cap\mathcal F\cap\mathcal W}, \ \overline{\mathbb E}, \ \overline{\mathfrak s}\bigr)$ is not canonically triangulated. Between this extriangulated structure and the one arising from the triangulation, the intermediate extriangulated structures are in one-to-one correspondence with Serre subcategories of $\mathrm{fp}((\frac{\mathcal P(\mathcal C\cap\mathcal F)}{\mathcal C\cap\mathcal F\cap\mathcal W})^{\mathrm{op}}, \mathrm{Ab})$.
Read the full abstract
Statistical Methodology
Random mixtures in Bayes Hilbert spaces
We present a framework for the analysis and unmixing of random density mixtures in the Bayes Hilbert space. General identifiability results for mixtures in Hilbert spaces are established and applied to the Bayes Hilbert space setting. Building on these results, we propose a penalised maximum likelihood approach for the unmixing of Bayes Hilbert mixtures aimed at recovering the statistically space-efficient representation, together with a computationally efficient coordinate-wise maximisation algorithm for its implementation. The methodology is illustrated through a hyperspectral data application, where observations can be naturally embedded in the Bayes Hilbert space and analyzed in terms of distributional shape rather than amplitude. A complementary simulation study demonstrates the interpretability and practical performance of the proposed approach.
Read the full abstract
Information Retrieval
EPIC: Explicit Posterior Item Conditioning for Semantic ID Diffusion Recommendation
Semantic ID (SID) generative recommendation predicts the next item by generating a short tuple of discrete tokens. Recent masked-diffusion methods improve this process through bidirectional context and flexible decoding, yet recommendation ultimately requires selecting among complete catalog items. At each denoising step, a partial SID can correspond to multiple feasible items, while existing methods primarily reason through position-wise token predictions. We propose Explicit Posterior Item Conditioning (EPIC), which introduces explicit item-level competition into SID denoising. EPIC constructs a personalized posterior over feasible candidate items using the current generation context and the user's recent interactions, then projects this distribution back to unresolved SID positions to guide subsequent token decisions. The pretrained backbone remains frozen and requires no additional decoder forward pass. Experiments on four Amazon benchmarks show consistent improvements over strong baselines, while diagnostic analyses indicate that the gains primarily arise from personalized transition evidence that preserves promising item hypotheses during denoising.
Read the full abstract
Quantitative Methods
The Identification of Biological Stains at Crime Scenes: A Promising Role for Proteomics and Machine Learning
Forensic body fluid identification is crucial for reconstructing crime scene events. While DNA analysis provides individualization, it lacks information about the fluid's origin. We developed and evaluated three complementary proteomic approaches using LC-HRMS/MS to identify blood, saliva, semen, urine, and vaginal fluid, including complex mixtures. The first method utilized fluid-specific peptide biomarkers, achieving high accuracy for pure fluids. The second employed peptide abundance ratios, demonstrating effectiveness in body fluid mixtures. The third, a machine learning model using Classifier Chain Random Forest, achieved 100% accuracy for pure fluids and promising results for mixtures. Our results revealed the complementarity of different tests, with the peptide-specific biomarker and machine-learning approaches being the most robust. This study demonstrates the potential of proteomics for comprehensive body fluid identification, offering valuable tools for forensic investigations.
Read the full abstract
Computer Vision
Neural Video Compression Based on Deformable Temporal Alignment and Difference-aware Fusion
In conditional coding-based neural video compression, the quality of temporal context directly affects compression per- formance. Existing methods mostly construct context from prop- agated reference features, but they are vulnerable to motion esti- mation and local alignment errors in regions with complex mo- tion, occlusion, and high-frequency textures, resulting in inaccu- rate temporal information. To address this issue, this paper pro- poses a method combining deformable temporal alignment and difference-aware spatial selective fusion. A Context-aware Tem- poral Alignment Module is used to generate complementary tem- poral context, while a Difference-aware Spatial Selective Fusion module adaptively selects reliable temporal information and sup- presses misalignment. Experiments show that the proposed method achieves certain rate-distortion performance improve- ment over DCVC-DC.
Read the full abstract
Strongly Correlated Electrons
Non-Resonant Impulsively Stimulated Raman Scattering by a Terahertz Field: a Case Study of 1T-TaS2
Time-domain ultrafast and nonlinear terahertz spectroscopy techniques are recently applied to many condensed matter systems for investigating their collective excitations. In centrosymmetric systems, these collective modes are typically Raman-active and therefore do not couple directly to the terahertz electric field. The mechanism by which light-matter interaction realizes in these studies has not been explicitly discussed in detail. In this work, we perform terahertz pump - optical probe and terahertz third harmonic generation investigations on 1T-TaS2, a material exhibiting a rich charge-density-wave (CDW) phase diagram including the commensurate, nearly-commensurate and incommensurate CDW phases. The transition between these distinct states leaves a clear signature on the dynamical Raman response. We investigate how the Raman-active phonons couple to a broadband monocycle terahertz field as well as a narrowband multicycle terahertz field. Our results indicate that a modified impulsively stimulated Raman scattering mechanism involving two-photon absorption, also known as non-resonant Raman scattering, underlies the coherent excitation and observation of the lattice modes. These results are relevant for future spectroscopy investigation and coherent control of collective modes using low-energy terahertz field as well as cavity electrodynamical dressing of solids.
Read the full abstract
Materials Science
Orbital-Free DFT-Assisted Machine-Learned Molecular Dynamics for Electric-Field-Driven Ionic Transport
We propose an orbital-free density functional theory (OFDFT)-assisted machine-learned molecular dynamics method in which field-independent interatomic forces are evaluated using a machine-learned potential and atomic charges that depend on the local environment are obtained from OFDFT calculations. The atomic charges obtained by Bader partitioning of the OFDFT electron density are multiplied by the electric-field vector and added to the forces from the machine-learned potential. This approach enables molecular dynamics simulations under an electric field at a lower computational cost than Kohn-Sham DFT (KSDFT)-based molecular dynamics. As a proof of concept, the method was applied to beta-Li3PS4 bulk and an S8/Li3PS4 heterostructure under periodic boundary conditions. For Li3PS4 bulk, OFDFT yielded Li charges and a total charge of the PS4 unit consistent with those obtained using KSDFT. In the heterostructure, Li ions migrated from the Li3PS4 region into the S-rich region within a simulation time of approximately 170 ps. Accompanying this migration, the mean Bader charge of the S atoms that initially formed S8 rings changed from nearly neutral to negative. A comparison with KSDFT for a representative interfacial structure also confirmed that the negative charging of the S atoms upon the arrival of Li and the magnitude of the Li charges were qualitatively reproduced. These results demonstrate the possibility of treating field-driven ionic transport and changes in interfacial charge states using a universal machine-learned potential without training an additional material-specific charge-prediction model.
Read the full abstract
Materials Science
Electron-phonon coupled hydrodynamics in semimetal TaAs_2
Hydrodynamic corrections to diffusive transport can arise when momentum-conserving collisions between quasiparticles become prominent, and they have been documented for both electrons and phonons. An emerging frontier topic is coupled electron-phonon (e-ph) hydrodynamics. Here, through electrical and thermal transport measurements on TaAs2 crystals with different impurity levels, we document the emergence of an e-ph bifluid in the temperature window of 5 to 15 K. Within this range, the lattice thermal conductivity exhibits a faster-than-T^3 temperature dependence, as a consequence of non-monotonic and purity-dependent phonon mean free paths, a signature of phonon Poiseuille flow. However, strong e-ph coupling impedes the emergence of a ballistic regime. This is corroborated by the observation of quantum oscillations in the lattice thermal conductivity. Prominent phonon-mediated momentum exchange between electrons amplifies the violation of the Wiedemann-Franz law and yields a two-order-of-magnitude discrepancy between quantum and transport lifetimes, a signature of electron hydrodynamics in semimetals. Our results imply that in semimetals with optimized e-ph coupling, thanks to matching between the cryogenic phonon wave?length and the Fermi wavelength, momentum and energy flow between the electron and phonon reservoirs as frequently as within each reservoir.
Read the full abstract
Computer Vision
Residual Optimal Transport-Based Experts Collaboration Towards Modality-Aware Infrared-Visible Object Detection
Infrared-visible object detection (IVOD) integrates complementary evidence from visible and infrared sensors for reliable perception in challenging scenes. In practice, sensors may fail or drop frames, leaving one modality unavailable or intermittent. Existing methods for IVOD assume both modalities are always present, and fixed fusion collapses when one stream is missing. Furthermore, it remains a critical challenge to reliably estimate semantic correlation across heterogeneous modalities, especially under spectral distribution discrepancy. We present FlexibleFusion, a unified and adaptive method that flexibly allocates integration pathways and fusion strength, operating seamlessly across complete and missing-modality regimes. At its core, the Modality-Aware Experts Collaboration (MAEC) mechanism selectively activates and aggregates cross-modal or intra-modal expert pathways. It allows cross-modal fusion when full modalities are available and falls back to self-fusion under missing conditions. Additionally, we design Residual Self-Paced Entropic Optimal Transport (RSPEOT) to align heterogeneous feature distributions from a transport perspective. Instead of relying on the fixed sparsity coefficient in standard entropic optimal transport (EOT), RSPEOT introduces a residual-driven self-paced update that prioritizes reliable matches and progressively refines harder ones. This design alleviates the additional optimization burden of standard EOT while preserving reliable semantic alignment. Comprehensive experiments under complete and missing-modality protocols show consistent performance across arbitrary modality configurations. Code will be released upon publication.
Read the full abstract
Artificial Intelligence
What Matters for Aggressive Decoding-Time KV Eviction? Temporal Aggregation and Ranking Preservation
Decoding-time KV cache compression research focuses heavily on designing better token scoring functions, while the temporal rule that aggregates scores across decode steps is often treated as an implementation detail. Under aggressive KV compression, we find that exponential-moving-average (EMA) aggregation makes approximately order-preserving scorer modifications largely indistinguishable at the eviction-set level. Value-norm and entropy variants remain highly correlated with attention and produce nearly unchanged retention sets, whereas KeyDiff, key norm, recency, and a learned scorer alter the ranking and degrade substantially. We associate this stability with the evaluated aggregation, which couples layer weighting and temporal retention. Building on this observation, we introduce InertiaKV, an EMA-based decoding-time eviction method, and InertiaKV-Lazy, its periodic-refresh variant, which yields 1.34-1.46x decode throughput relative to full refresh InertiaKV. We also study Score-Free decoding as a separate empirical operating point: it scores the full context once at the first decode step, freezes that ranking, and incurs an average quality change of +0.03 while removing all subsequent scoring. Across six open-weight backbones and the LongBench, LongBench-v2, and RULER benchmarks, the results identify temporal aggregation and ranking preservation as distinct, consequential design factors; they do not imply that scoring quality is irrelevant in general.
Read the full abstract
Category Theory
Stratification of Artin motives over local fields
Let k be a field of characteristic p>0. We prove that for every p-decomposition group P, the tensor-triangulated category \operatorname{DPerm}(P;k) is stratified and its Balmer spectrum is generically noetherian. As an arithmetic application, we show that for a nonarchimedean local field F with residue characteristic \ell, the category \(\operatorname{DAM}(F;k)\) of derived Artin motives is stratified if and only if \ell\neq p. In the stratified case its Balmer spectrum is generically noetherian; consequently, \operatorname{DAM}(F;k) satisfies the telescope conjecture.
Read the full abstract
Strongly Correlated Electrons
Competing lattice structures induced by Sn substitution in CsV$_3$Sb$_5$
Understanding the effect of chemical substitution on competing phases of kagome metals is crucial for disentangling the interplay between local structural distortions and electronic instabilities. In \cvs, Sn substitution strongly modifies the electronic phase diagram, yet the microscopic mechanism driving this remains unclear. Here, we combine $^{121}$Sb nuclear quadrupole resonance (NQR) measurements and density functional theory calculations to investigate the atomic-scale effects of Sn substitution in CsV$_3$Sb$_{5-x}$Sn$_x$. At low Sn concentrations, the observed satellite NQR peaks exhibit signatures of local structural distortion induced by Sn substitution, qualitatively consistent with our computational analysis. These impurity-induced features persist across the entire experimentally investigated doping range, up to $x$ = 0.65, and remain observable up to room temperature. For $x=1$, the fully doped idealized case, the estimated dynamical instabilities of the kagome lattice suggest the stabilization of two nearly energy-degenerate equilibrium structures characterized by V-trimers, distinguished by a zero- or $\pi$-phase shift between adjacent layers along the $c$-axis. Together, these results show that Sn substitution drives a complex interplay between local impurity-induced distortions and competing structural instabilities in vanadium-based kagome compounds.
Read the full abstract
Combinatorics
On graphs with equal domination and total domination numbers
For a graph $G$ without isolated vertices, $\gamma(G)\le\gamma_t(G)\le 2\gamma(G)$. While graphs attaining $\gamma(G)=\gamma_t(G)$ have been studied extensively, a complete structural description in the smallest nontrivial case $\gamma(G)=2$ has remained open. We resolve this case according to girth. When $g(G)\ne 3$, we show $\gamma_t(G)=2$ forces $G$ bipartite, give an exact degree-sum criterion for this equality, and show $\gamma_t(G)\in\{2,4\}$ under the additional hypothesis $\delta(G)\ge 2$. When $g(G)=3$, we use Golumbic's vertex-multiplication operation together with known classifications of graphs of rank $2$ through $5$ to completely list the families satisfying $\gamma(G)=\gamma_t(G)=2$. As an application, we show that every graph in the extremal family of diameter-two, dominating-vertex-free graphs identified by Erdős and Rényi and classified by Henning and Southey satisfies $\gamma_t(G)\in\{3,6\}$, so $\gamma=\gamma_t=2$ never occurs there. Together these results give a full structural dictionary translating $\gamma_t(G)=2$ into concrete, checkable graph-theoretic properties.
Read the full abstract
Computation and Language
Lost in Reordering: Structural Sensitivity of Multilingual LLMs under Semantics-Preserving Perturbations
Large Language Models (LLMs) demonstrate strong multilingual reasoning performance, yet their robustness to semantics-preserving structural variation remains underexplored, particularly for relatively free word-order languages. We investigate the structural sensitivity of multilingual LLMs using two linguistically grounded perturbation settings in Hindi and Malayalam: constrained constituent reordering and active-passive voice transformation. We introduce a benchmark dataset IndicReStruct, with two variants, GSM8K-Reordered and GSM8K-Voice, constructed from GSM8K while preserving semantic meaning. Across six state-of-the-art LLMs and multiple prompting strategies, we observe consistent and significant degradation in mathematical reasoning performance under structurally perturbed inputs. To further understand these failures, we perform qualitative error analysis and mechanistic interpretability experiments using residual-stream activation patching. Our analyses show that reasoning failures frequently arise from disruptions in entity-quantity alignment and that intermediate transformer layers contribute most strongly toward reasoning restoration. Overall, our findings suggest that current multilingual LLMs remain highly sensitive to surface syntactic realization and lack robust compositional invariance under structurally different but semantically equivalent inputs.