Peer Revenue Play credits. Real papers.

Papers

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

All of arXiv

Wednesday 9 September

Audio and Speech Processing

Candor-LR: A Dyadic Conversational Dataset for Audio-Visual Speech Recognition

Current audio-visual speech recognition (AVSR) benchmarks, like LRS3, rely heavily on clean, scripted and rehearsed speech. They fail to reflect the complexity of natural conversation, which involves overlapping speech, spontaneous turn-taking, unscripted vocabulary and variable acoustic conditions. To shift the field toward realistic dialogue, we introduce Candor-LR, a conversational benchmark derived from the CANDOR corpus of 1,656 natural dyadic videoconferences. Our custom data preparation pipeline yields 713.5, 10.1, and 60.1 hours of training, validation, and test data, respectively. Evaluating pretrained AVSR models on Candor-LR reveals that audio-only accuracy drops sharply compared to LRS3, but visual cues compensate effectively, driving much larger performance gains on Candor-LR than on LRS3. Furthermore, training on this corpus significantly improves cross-domain robustness under both clean and noisy conditions, as its realistic conversational data captures broader audio-video features. We open-source our pipeline to ensure reproducibility, establishing Candor-LR as a challenging benchmark for conversational AVSR.
Read the full abstract

All authors Rishabh Jain, Aristeidis Papadopoulos, Zhaofeng Lin, Naomi Harte

Soft Condensed Matter

A Sharpened Entropy Principle for Two-Fluid Polymer Thermodynamics

A compressible, non-isothermal dilute polymer solution is formulated as a Class-II binary mixture with separate solvent and polymer mass and momentum balances and with total energy and entropy balances for the complete mixture. The entropy exploitation is sharpened by a balance-anchoring axiom introduced here: in a fixed balance representative, each selected binary dissipative product contains a constitutive factor from an unclosed balance flux or source. Polymer configuration is resolved first by a connector distribution and then by its conformation tensor. Population kinematics fix polymer transport and, by moments, yield a two-velocity upper-convected rate; objectivity verifies covariance rather than selecting it. Within the total entropy balance, configurational transport and deformation powers cancel their chemical-potential and elastic partial-stress counterparts when deformation and stress-decomposition weights match. A Gordon–Schowalter test independently requires the affine upper-convected choice for the stated dumbbell free energy and Kramers stress unless an additional reversible channel is supplied. Coordinated stress–interaction changes shift the local entropy flux/production pair by a divergence, exposing representation dependence of local mechanism-wise production. An entropy-invariant Class-II-to-Class-I reduction selects a descendant entropy flux preserving the parent production and yields thermo-chemical, configurational-stress and partial-viscous-stress diffusion terms. The omitted quadratic relative-inertia flux is paired with relative kinetic-energy storage and transport and is a reversible truncation, not missing entropy production. The resulting compressible non-isothermal Hookean stress and temperature equations reduce to Oldroyd-B/UCM only after one-velocity, incompressible and isothermal limits.
Read the full abstract

Audio and Speech Processing

Teacher-Free Self-Distilled Consistency Trajectory Learning for Fast Speech Enhancement

Consistency trajectory models offer a route to fast, high- quality speech enhancement, collapsing the many reverse steps of diffusion-based enhancers into a handful. When instantiated on a Schrödinger bridge (SB), which pins the generative process to fixed clean and noisy endpoints, exist- ing consistency-trajectory enhancers (SBCTMs) still require a pretrained teacher to supply trajectory supervision, which raises training cost and ties the final quality to that of the teacher. We propose a teacher-free, self-distilled consistency- trajectory framework that removes the external teacher result- ing in a 5X reduction in per epoch training time. Our model is trained with a three-stage curriculum of clean speech pre- diction, a self-distilled shortcut objective, and perceptual fine-tuning with a multi-resolution short-time Fourier trans- form (MR-STFT) loss. Using the same NCSN++ backbone as SBCTM, our model attains a wide-band PESQ of 3.01, ES- TOI 0.87, and SI-SDR 19.07 dB on VoiceBank+DEMAND compared to 3.57, 0.87 and 12.8 dB for the teacher based model. Further, we find that a geometric schedule at low reverse step count maximizes perceptual quality, while a higher-step uniform schedule favors signal fidelity.
Read the full abstract

Statistical Mechanics

The Arithmetic of Spectra: Factorization, Statistics, and Symmetric Functions

We study factorizations of the single-particle spectrum of non-interacting quantum systems and their consequences for many-particle statistics. Representing the single-particle partition function by a spectral alphabet, tensor factorizations become multiplicative factorizations of the alphabet, which can be lifted to canonical Bose and Fermi partition functions using standard symmetric-function and $\lambda$-ring identities. We show how product spectra arise from different factorizations, how antisymmetrization can be assigned across an odd number of factors, and which tensor factorizations are compatible with a fixed spectrum. The paper therefore studies spectral factorization and provides a consolidated combinatorial framework for canonical Bose and Fermi partition functions.
Read the full abstract

General Relativity and Quantum Cosmology

Field quantization in rotating frames: coordinate covariance and the circular-detector response

We provide a unified analysis of scalar-field quantization and detector response in relativistic rotating frames, comparing rigid and Trocheris-Takeno (TT) descriptions in unbounded Minkowski spacetime. We distinguish three questions that are often conflated: (i) how a fixed quantum state is represented in rotating coordinates, (ii) whether rotating time evolution selects a global ground state, and (iii) how a rotating detector responds to the vacuum. Coordinate-transformed rigid and TT modes span the same positive-frequency subspace as inertial modes: their Bogoliubov beta coefficients vanish and their mode sums reproduce the Minkowski Wightman function. Neither rotating time flow selects a global scalar ground state, although for different reasons: the rigid generator is not globally timelike and is unbounded below, whereas TT time translation is not a Killing symmetry. Nevertheless, a circular Unruh-DeWitt detector in the Minkowski vacuum has a stationary, nonzero response. We relate this response to negative corotating frequencies, evaluate it after an explicit Hadamard subtraction, and show that it is nonthermal. Thus vanishing Bogoliubov mixing, the absence of a symmetry-selected rotating vacuum, and detector excitation are mutually consistent.
Read the full abstract

Pattern Formation and Solitons

Fast Synergetic Simulation to Study Slow Evolution of Soliton Patterns in Optical Resonators

Complex patterns in physical and biological systems often emerge through slow collective dynamics governed by a small number of key variables. In nonlinear optical resonators, dissipative Kerr solitons provide an important example, where interactions between well-separated solitons can evolve over timescales far longer than the characteristic loss and gain timescales. Direct numerical simulation of these dynamics is challenging because stiffness forces conventional methods to resolve many rapidly damped degrees-of-freedom with very small time steps. We present a numerical scheme, the synergetic method, that eliminates these rapidly damped degrees-of-freedom and retains the slowly evolving modes, enabling time steps many orders of magnitude larger than those used in conventional approaches. Applied to soliton molecules in driven Kerr cavities, the method achieves speedups of $10^3$ to $10^5$ while capturing dynamics on laboratory timescales. We use it to model the full interaction dynamics of a three-soliton molecule and the evolution of an eight-soliton molecule. The approach provides an efficient framework for studying slow pattern formation in nonlinear systems with widely separated timescales.
Read the full abstract

All authors Sanzida Akter, Pradyoth Shandilya, Logan Courtright, Amir Leshem, Giuseppe D'Aguanno, Omri Gat, Curtis R. Menyuk

Systems and Control

Economic Evaluation of V2G-Enabled Fast Charging Stations Under Endogenous EV Adoption Dynamics

Building fast charging stations (FCSs) is crucial for transportation electrification, but there exists an indirect network effect: while the increasing number of electric vehicles (EVs) decides the FCS capacity expansion, the spatial locations of these facilities strongly influence drivers' willingness to adopt EVs. Ignoring this interaction can lead to bad capital investments and exacerbate power grid vulnerabilities during tidal traffic peaks. Therefore, we explicitly model the EV adoption dynamics as decision-dependent uncertainties (DDUs) in a new multi-period collaborative planning framework. This framework evaluates the economic viability of V2G-enabled FCSs across both transportation and distribution networks. To simplify the complex calculation, we introduce an aggregated fleet virtual battery model to catch macroscopic vehicle-to-grid (V2G) flexibility. This successfully circumvents the dimension curse inherent in tracking microscopic state-of-charge. To further guarantee calculation speed, the nonlinear infrastructure exposure is transformed into a mixed-integer program by using Special Ordered Set type 2 (SOS2) variables and Second-Order Cone Programming (SOCP) relaxations for grid limits. Finally, numerical studies on a coupled Sioux Falls and IEEE 33-bus testbed prove that our framework achieves superior expected social welfare. Also, macroscopic V2G aggregation is highlighted for its capability to mitigate distribution grid congestion penalties.
Read the full abstract

All authors Mingjian Tuo, Jie Zhou, Yao Yan, Cunzhi Zhao, Long Wang, Mulan Zhang

Computer Vision

Enhanced Deformable Convolution with Center-invariant Offset and Edge-aware Mask

Deformable convolution networks have recently become popular for many computer vision tasks, especially for semantic segmentation, because of their exceptional capabilities in dynamic spatial modeling. However, due to the dense deformable offsets and the lack of longer-range dependencies, they can not fully adopt proper and precise deformations for feature representations. To tackle the issues, in this paper, we propose Enhanced Deformable ConvNets (EDCN) for semantic segmentation. Specifically, a novel Enhanced Deformable Convolution (EDC) is exploited in the decoder, which integrates the Center-invariant Offset Module (COM) and Edge-aware Mask Module (EMM). The COM employs larger kernels and eliminates deformations at the kernel center, obtaining offsets that are more in line with the target from richer spatial information. Concurrently, the EMM obtains the significance of image content via Sobel edge detection, then selectively applies deformations based on the content significance, minimizing unnecessary deformations associated with relatively less important information, thereby avoiding impact from less informative regions. Experiments show that EDC outperforms state-of-the-art deformable convolution variants, including Deformable ConvNets V1-V4 and Entire Deformable ConvNets, across mainstream segmentation datasets with various decoder settings. Moreover, ablation studies confirm the effectiveness of each component. In addition, visualizations illustrate that EDC enhances spatial adaptation and target focus. We further analyze the extendibility of EDC to larger kernels on the image classification benchmark. Code will be publicly released.
Read the full abstract

All authors Yixiao Li, Xiaoyuan Yang, Jin Jiang, Minghao Zou, Guanghui Yue, Baoquan Zhao, Jun Liu, Wei Zhou

Applied Physics

Making the Graph Laplacian Physical: Multiscale Coarse-Graining in Electrical Oscillator Networks

Spectral gaps are widely interpreted as signatures of collective organization, yet it is rarely clear whether the resulting modes correspond to physical variables or merely to convenient mathematical coordinates. Here we show that, in resistor-coupled LC networks, the graph Laplacian defines an experimentally accessible hierarchy of electrical descriptions. Its eigenvectors become measurable voltage patterns, while its eigenvalues quantify the coupling-induced resistive damping of those patterns. When coupling is strong within regions and weak between them, microscopic voltages collapse into nested collective variables. A network of 1002 resonators reduces first to 42 regional voltages and then to four geometry-controlled global modes. Summing the microscopic Kirchhoff equations produces a directly constructible 42-node RLC circuit, without parameter fitting, that reproduces regional dynamics with 1.4 to 4.6% global RMS error. Finer modal coordinates improve accuracy but no longer correspond generally to simple resistor graphs. Robustness tests show that the regional level survives substantial conductance disorder and internal rewiring, but weakens when perturbations impair mixing within a region; changes in interregional connectivity selectively disrupt the global level. Thus the Laplacian spectrum becomes a physical design principle: it identifies which collective electrical variables emerge, how they can be constructed, and when they provide an adequate reduced description.
Read the full abstract

Human-Computer Interaction

MOONWALK: Mediating Operations with Intent-Evidence-Action Alignment Across Junior-Supervisor Review Workflows in Animation/VFX Pre-Production

Animation and VFX pre-production review requires teams to translate loosely specified creative intent–briefs, evolving specifications, heterogeneous references, and verbal decisions–into revisions that junior artists can execute without repeated clarification. In practice, criteria drift across iterations, review judgments lose their evidential basis, and the reasoning behind a request rarely survives the senior-junior handoff. We contribute a design framework for intent-evidence-action alignment: intent is articulated into a shared project record, judgments are anchored to grounded evidence, and authorized decisions are converted into clear revision tasks tied directly to reference notes. We instantiate this framework in MOONWALK, a professional pre-production review system comprising a shared intent record, reference/specification anchoring, structured work-in-progress comparison, and supervisor-authorized action planning. In this workflow, AI handles administrative coordination–flagging missing context and organizing notes–while artists retain full creative direction. An in-studio study with professional practitioners compares MOONWALK with a chat-only (chatbot) interface using matched production materials, while participants' existing workflows provide a retrospective ecological baseline. Results indicate stronger intent alignment, decision traceability, and checklist executability, while also showing that aesthetic authority and final prioritization must remain with practitioners. The evaluation establishes the value of the integrated structured workflow over unstructured conversational AI chatbot. Code: https://github.com/Akinesia112/Moonwalk/tree/english-version
Read the full abstract

All authors Shih-Yu Lai, Wen-Fan Wang, Sai Ling, Shaune Jan, Bing-Yu Chen, Xiang Anthony Chen

Instrumentation and Methods for Astrophysics

Wide Field Localization Using Fixed Modulation Collimators

Rapid, precise localization of gamma-ray bursts (GRBs) requires continuous wide-field imaging at hard X-ray and soft gamma-ray energies. We develop a fixed (non-rotating) modulation-collimator architecture to address this challenge. A bigrid collimator, a pair of aligned, periodic slit grids in front of a simple, non-imaging photon-counting detector, transmits an off-axis source through a triangular fringe pattern whose local angular position, the source phase, is recovered from simultaneous count rates in several grid pairs at fixed relative phase offsets, rather than by mechanically rotating a single grid pair as in the classical rotating modulation collimator. Four-phase demodulation of these count rates robustly cancels detector background and grid leakage without requiring rotation. Because the fringe pattern repeats across the sky, a phase measurement corresponds to many candidate source positions; we resolve this ambiguity over an arbitrarily wide field of view using a cascade of progressively coarser-pitched grid stages, generalizing the two-grid vernier-scale technique originally used to localize Sco X-1. This vernier cascade technique reduces the required photon counts and grid-calibration accuracy required to achieve fine angular resolution over a broad field of view. Generalized to two dimensions with orthogonal cascades, the resulting architecture achieves wide-field localization with no moving parts, no time modulation, and only simple photon-counting detectors, a promising candidate for simple, inexpensive, continuous-monitoring instruments supporting time-domain and multi-messenger astrophysics.
Read the full abstract

High Energy Physics (Phenomenology)

Quantifying Information Hierarchy for Neutrino Oscillation Parameters at JUNO

Since neutrinos are quantum systems inherently, the precision with which oscillation parameters can be estimated ultimately depends on how much information about these parameters is encoded in the neutrino state and how efficiently that information can be extracted through measurement. In this work, we quantify how information encoded in reactor antineutrino states flows through the measurement process to the events observed at the detector, using quantum and classical Fisher information. We establish the information ladder for JUNO, revealing that the loss of precision across different information levels is strongly parameter dependent. We demonstrate that the JUNO configuration approaches the optimal statistical limit for the oscillation parameters of the solar sector, while information on $\theta_{13}$ and $\Delta m_{31}^{2}$ is significantly degraded by the measurement strategy and detector effects. Despite this information loss, the remaining information is sufficient for JUNO to achieve sub-percent precision on $\Delta m_{31}^{2}$ within six years.
Read the full abstract

All authors Yu-han Shu, Neetu Raj Singh Chundawat, Luis A. Delgadillo, Yu-Feng Li

High Energy Physics (Phenomenology)

Searching for New Physics with Reinforcement Learning

Finding new physics (NP) is the most important problem in particle physics today. Studying “anomalies”, i.e., measurements of low-energy observables whose values disagree with the predictions of the Standard Model (SM), is a powerful search strategy. The SM Effective Field Theory (SMEFT) provides a general model-independent framework for parameterizing NP; it is natural to try to find the SMEFT operator(s) that can explain such anomalies. This is a challenging task because (i) the number of SMEFT operators is enormous, and (ii) at loop level there are very complicated correlations among the operators. Analyses by humans typically rely on phenomenological intuition to decide which operators are relevant. This is often biased and does not explore the complete SMEFT operator space. Interestingly, reinforcement learning (RL) techniques excel at tasks that require decision making to achieve their goals. In this paper, we introduce an RL method that can be used to find the SMEFT operators that explain any anomalies. We test it on the CDF $W$-mass anomaly, and show that it reproduces (and improves upon) known results. We then consider a far more complicated situation with multiple anomalies and show that, even here, this method is able to find the SMEFT operators that explain the data. Our RL method can therefore be used to efficiently search for NP at the level of SMEFT.
Read the full abstract

Superconductivity

A First-Principles Multiscale Framework for Topological Superconductivity

A microscopic understanding of topological superconductivity (TSC) in real materials requires a materials-informed approach that integrates first-principles electronic structure, superconductivity, and topology within a unified computational framework. Here, we develop such an approach by combining density functional theory, Wannier-based low-energy Hamiltonians, Bogoliubov-de Gennes theory, and Matsubara Green's-function-based Chern number calculations performed directly on realistic multiorbital superconducting Hamiltonians. We apply this framework to bulk-like and monolayer FeTeSe and to FeSe/GaAs heterostructures, enabling a unified investigation of both intrinsic and proximity-induced topological superconductivity. We identify key electronic-structure ingredients that promote robust TSC, including Rashba-active states near the Fermi level, strong induced superconducting pairing, and substantial orbital hybridization between spin-orbit-active and superconducting sectors. Guided by these design principles, we predict multiple topological superconducting phases in Fe-based materials and demonstrate that FeSe/GaAs heterostructures are particularly promising, exhibiting topological transitions at experimentally accessible chemical potentials and low Zeeman fields. Complementing the theoretical predictions, we demonstrate the growth and structural characterization of FeSe/GaAs heterostructures, establishing the experimental feasibility of the proposed materials platform. Our results provide a quantitative route for engineering and screening candidate Majorana materials and heterostructures directly from realistic electronic structures and establish a foundation for future materials-by-design approaches to topological superconductivity.
Read the full abstract

All authors Christopher L. Jacobs, Baishali Mandal, Taehwan Jung, Purna P. Paudel, Jason Kawasaki, Tudor D. Stanescu, Subhasish Mandal

General Economics

High Volume Low Complexity Surgical Hubs in England: Can They Improve Physician Productivity?

Whether organisational separation of elective and emergency care improves physician productivity remains an open question. Most existing evidence relies on cross-sectional comparisons or volume-based outcomes that cannot isolate efficiency gains from input expansion or provider selection. In contrast, this paper exploits the staggered rollout of NHS England's surgical hub programme, a set of dedicated ring-fenced elective facilities introduced as part of its Elective Recovery Plan, to provide causal estimates of the effect of elective-emergency separation on physician productivity. Using this rollout to implement a heterogeneity-robust difference-in-differences design, we estimate the impact of hub adoption on physician productivity measured as cost-weighted elective output per unit of salary-weighted physician input. We find that hub adoption increases physician productivity by 14.5% relative to the no-hub counterfactual, accompanied by a 10-day (7.6%) reduction in average patient waiting times. The size of these gains depends on how completely elective care is insulated from emergency pathways. Standalone hubs situated on dedicated elective-only sites achieve relatively larger productivity gains than integrated hubs located within main acute hospital sites. Moreover, operating multiple hubs yields gains more than double the overall average, whereas a single hub shows no statistically significant effect. For health systems seeking to raise physician productivity and address elective backlogs, these findings suggest that how separation is implemented, not simply whether it is adopted, shapes the productivity gains it delivers.
Read the full abstract

All authors Zecharias Anteneh, Adriana Castelli, Peter Sivey, Andrew Street, Jinglin Wen, Joy Adamson

Algebraic Geometry

An orbifold formula for algebraic stacks

We study motivic integration on varieties that are birational to smooth algebraic stacks and prove a generalization of the orbifold formula for varieties with finite quotient singularities. This leads in particular to a new way of computing stringy E-functions and we also give applications to the study of klt-singularities. Along the way we study Cluckers-Loeser motivic integration on algebraic stacks, which might be of independent interest. The proof relies furthermore on ideas from Hrushovski-Kazhdan integration and Bruhat-Tits theory.
Read the full abstract

Atmospheric and Oceanic Physics

Predicting tipping points: The many shades of non-equilibrium and catch-22s of early-warning

The potential of crossing climate tipping points has reached the attention of many researchers and the general public. On the one hand, the basis for this concern is strengthening, with simulations showing abrupt transitions that may occur even for moderate greenhouse gas emission scenarios. On the other hand, our understanding of what constitutes such transitions mathematically is becoming more nuanced. This leads to challenges for the fidelity of early-warning signals (EWS), which accompany bifurcations in systems that closely track a slowly changing steady state. Different kinds of out-of-equilibrium dynamics in a rapidly changing climate, as well as chaos, blur the notion of a critical tipping threshold and imply limits in predictability. Using EWS to detect loss of local stability from data requires careful evaluation in the high-dimensional system, and may have limited predictive power in a highly multistable climate, where there is more than a single, well-known alternative state. This calls for new methods using non-equilibrium statistical mechanics and dynamical systems to efficiently probe the global stability properties of climate models and observations.
Read the full abstract

Robotics

Data-Driven Risk Fields for Safer End-to-End Autonomous Driving

Safety is a fundamental requirement for autonomous driving, yet existing end-to-end driving models still lack explicit risk-aware learning capacities. Existing rule-based risk models provide interpretable safety priors, yet their absolute risk scores depend on handcrafted functions, coefficients, and thresholds. Learning-based risk representations reduce part of this manual design, but their supervision often relies on occupancy-derived labels or heuristic cost values, which may not capture ego-conditioned planning risk. In this paper, we propose DRiF, a data-driven risk-field framework for safer end-to-end autonomous driving. DRiF learns a shared BEV feature with static map segmentation, dynamic risk prediction, and vehicle planning. For dynamic risk learning, DRiF converts rule-based safety priors into pairwise risk labels, and trains the risk field to preserve relative risk ordering instead of regressing handcrafted absolute scores. Experiments on Bench2Drive show that DRiF achieves competitive overall performance, with consistent improvements in driving score, success rate, and collision-related metrics. These results establish relative risk supervision as an effective way to connect explicit safety structure with end-to-end planning. The data and code will be publicly available.
Read the full abstract

All authors Yuanxin Tian, Zhiyuan Liu, Jinhao Li, Liangfan Zhu, Shuai Wang, Heye Huang, Qingwen Meng, Fang Zhang, Liuzhu Tong, Zhenhua Xu, Wenhao Yu, Jianqiang Wang

Computer Vision

Shape-guided Gaussian Splatting for Sparse-View X-ray 3D Reconstruction

Sparse-view X-ray 3D reconstruction is essential for reducing radiation exposure, but recovering a density field from a handful of X-ray projections is severely ill-posed. Recently, 3D Gaussian Splatting has achieved state-of-the-art performance in sparse-view reconstruction by representing the volume using explicit, optimized primitives, but it requires dozens of projected views. With fewer views, reconstruction quality degrades severely since the explicit primitives are optimized freely without any anatomical information. Anatomical structures, in contrast, share similar geometry and density across a population. Their variations are bounded within a limited range that statistical shape models can capture. This paper proposes a shape-guided Gaussian splatting framework for sparse-view X-ray 3D reconstructions. Our contribution lies in driving Gaussian positions toward anatomically valid configurations, alongside atlas-based density regularization. Our method ensures anatomically consistent reconstruction and improves PSNR by 2.83 dB over a state-of-the-art Gaussian splatting baseline with as few as 5 views. Code Available: https://github.com/polyshape-lab/ShapeGuidedGaussian
Read the full abstract

All authors Pranav Poudel, Florence Dell'Aniello Picard, Nairouz Shehata, Frédéric Lavoie, Herve Lombaert

Earth and Planetary Astrophysics

From spirals to rings: dust dynamics in gravitoturbulent protoplanetary discs after late infall

Substructures appear to be a common trait of all extended protoplanetary discs. They are found notably in discs still surrounded by ambient, possibly infalling material. In this study, we revisit the observational signatures of gravitational instability (GI) in the dust and gas emission of protoplanetary discs. We focus on discs undergoing a short-lived episode of late infall that triggers spirals via GI, in order to investigate the long-term dynamics of millimetre-sized dust. We performed 2D hydrodynamical simulations modelling gas and millimetre-sized dust in a self-gravitating disc, with a simplified treatment of stellar irradiation and infall. Results were post-processed by 3D radiative transfer calculations. During infall, GI sets in and the disc develops a gravitoturbulent state characterised by the continuous formation of spirals. Shortly after infall ends, the millimetre-sized dust remains well coupled to the gas, and the dust emission exhibits large-scale spirals in both near-infrared scattered light and continuum emission. Even long after infall has ceased, two-armed spirals are persistently excited by GI in specific regions of the disc. Their dissipation entails the formation of multiple long-lasting pressure maxima, which can be viewed as vestiges of shocks induced by the spirals. They form persistent dust rings that are more or less axisymmetrical. Consequently, once late infall ends, a massive disc can gradually evolve from a disc with spirals in the continuum into one featuring multiple, potentially lopsided bright rings in the continuum. In near-infrared scattered light, the disc initially displays multiple spirals during infall or shortly after it ceases, before ultimately exhibiting multiple rings. The persistent excitation of spirals implies, however, that the residual line-of-sight velocity, derived from line emission, shows large-scale spirals across the disc.
Read the full abstract

All authors Clément Baruteau, Steven Rendon Restrepo, Gaylor Wafflard-Fernandez, Sebastián Pérez, Zhaohuan Zhu, Hans Baehr, Farzana Meru, Jean-François Gonzalez, Sergei Nayakshin, Sahl Rowther

Numerical Analysis

A quasi-interpolation method for finite element spaces

A combination of Taylor polynomial approximation and averaging results in a powerful quasi-interpolation technique that can be used to analyze the approximation properties of finite element spaces. With this particular approach, the notion of affine equivalent families is completely bypassed in the sense that the degrees of freedom are defined locally rather than being exported from a reference element. In particular, domains with curved boundaries can be handled in a natural manner. Finally, the ease of its application to a particular finite element makes it a valuable pedagogical tool.
Read the full abstract

Rings and Algebras

Amenability of Lie, Group and Hopf algebras

We propose to define amenability of a Lie algebra by the existence of almost-invariant finite-dimensional subcoalgebras of its universal enveloping algebra. More generally, a module coalgebra of a cocommutative Hopf algebra is amenable if it admits almost-invariant finite-dimensional subcoalgebras. We prove a coalgebraic rounding theorem: almost-invariant finite-dimensional subspaces can be replaced, without loss in the Følner constant, by almost-invariant finite-dimensional subcoalgebras. For Lie algebras this implies that our definition is equivalent to Elek's amenability of the left regular module of the universal enveloping algebra seen merely as an associative algebra. For groups this recovers the result that a group is amenable if and only if its group ring is algebraically amenable. It furthermore shows that, for every amenable group, all its nonzero modules are amenable, thus proving an assertion by Gromov. We prove that amenable Hopf algebras are closed under taking subalgebras, quotients, cleft extensions, and directed unions, and that every Hopf algebra locally of subexponential growth is amenable. We give examples of amenable Lie algebras which are not elementarily amenable. Finally, we show that amenability passes to the associated graded Hopf-module coalgebra.
Read the full abstract

Computer Vision

PACE: Perceived-Latency-Aware Cascading Service Routing and Filler Control for QoE-Efficient Retrieval-Augmented Dialogue Serving

We present the PACE, a framework for retrieval-augmented dialogue serving that formalizes Perceived Time-to-First-Response (PTFR) as a QoE objective and minimizes it under quality/cost constraints. Unlike prior work on cascaded routing, semantic caching, or adaptive retrieval, PACE jointly controls which answer source composes the response and what fills the waiting window. Deployed on a humanoid-robot sales service, it combines three mechanisms: a load-adaptive cascading router, a joint path-filler controller, and volatility-aware cache admission. On 75k CarQA requests, the cascade halves pure-LLM PTFR at P95 (0.29 vs 0.53s at c16). The adaptive controller reaches 0.41s P95, outperforming RAG by 2.4 times at high load with equal quality. The filler controller cuts calls by 94% with zero conflict. Volatility-aware admission reduces stale answers from 86% to 0%. A gating rule ensures the controller never worse than the baseline, with exposure bounded by one hold period. This is the first quantification of filler-answer conflict risk in deployed services.
Read the full abstract

All authors Lin Huang, Yujuan Tan, Weisheng Li, Lixiang Zeng, Kun Yang, Yongzong Wang, Suihan Xiao

Computer Vision

Beyond Weak Labels: Prompt-Guided Local Refinement for Weakly Supervised Water Segmentation in High-Resolution Multispectral Imagery

High-resolution water mapping supports environmental monitoring and related applications, but accurate pixel-level labels are difficult and costly to produce. Official hydrographic vectors provide scalable weak supervision, but they contain artifacts like boundary noise, temporal mismatch, and omissions of small water structures. We propose a two-stage framework for weakly supervised water segmentation in high resolution multispectral imagery. Stage 1 learns initial masks from rasterized vector pseudo-labels, and Stage 2 converts these masks into structured component-wise prompts for localized refinement. On a manually corrected validation set, refinement improves SegFormer-B0 from 0.9509 to 0.9535 IoU and U-Net from 0.9408 to 0.9486 IoU, with corresponding F1 gains from 0.9749 to 0.9762 and 0.9695 to 0.9736. It leads to sharper shorelines, reduced boundary spillover, and better thin-structure delineation. The results indicate that prompt-guided refinement can improve pseudo-label-based water segmentation by targeting local errors that are poorly captured by global training supervision.
Read the full abstract

All authors Muhammad Farhan Humayun, Mohammad Imangholiloo, Afifah Shah, Tomi Westerlund, Jukka Heikkonen

Numerical Analysis

Gradient-Enhanced Proximal Algorithms for Mean Field Planning on Surfaces

Mean field planning on a surface prescribes initial and terminal densities and minimizes a transport energy subject to the continuity equation. Proximal algorithms for this problem repeatedly solve a time–space Poisson equation, whose temporal derivative and surface gradient determine the density and momentum corrections. We study a gradient-enhanced approximation of this constraint projection using finite differences in time, surface finite elements in space, temporal polynomial preserving recovery, and spatial parametric polynomial preserving recovery. The same construction is incorporated into ISTA, FISTA, and Douglas–Rachford splitting. We distinguish the recovered update from an exact discrete projection and derive residual identities and conditional finite-iteration perturbation bounds that retain data, boundary, and linear-solver errors. Existing derivative-recovery estimates identify a higher-order contribution under suitable regularity and mesh assumptions; they do not by themselves establish convergence of the outer optimization iteration. Available numerical illustrations on the sphere and a more complicated algebraic surface are discussed together with the limits of the recorded refinement data.
Read the full abstract

Information Theory

Construction of Multi-sequences With High Nonlinear Complexity via Narrow Ray Class Fields

Nonlinear complexity is a fundamental criterion in the evaluation of pseudorandom sequences. The construction of multi-sequences with high nonlinear complexity is both theoretically and practically important in cryptography. Motivated by prior constructions of multi-sequences with high nonlinear complexity in [IEEE Trans. Inf. Theory, 60(10), 2014] and [IEEE Trans. Inf. Theory, 63(12), 2017], we provide a unified framework via narrow ray class fields and the cyclic descent introduced by Guruswami and Xing in [J. Combin. Theory Ser. A 129 (2015) ]. Then we can generate new multi-sequences with high nonlinear complexity over function fields with arbitrary genera.
Read the full abstract

Distributed Computing

Stencil Computation at the Intersection of AI and HPC

Tensor compilers such as TinyTC and OpenAI Triton were originally developed for AI workloads, but the same tiling and memory abstractions can be applied to implement efficient high-order stencils for scientific and industrial applications. We demonstrate this for an 8th-order, 25-point acoustic stencil with boundary conditions over an a demanding-sized grid, targeting GPGPUs, where we compare the hardware-specialized TinyTC implementation with a portable PyTorch/Triton implementation. The target platforms for evaluation include Intel B70, B580, GPU MAX 1550, NVIDIA A100/RTX6000 Blackwell/H100, and AMD MI325x. For instance, on Battlemage B580 TinyTC reaches 15.6 Gpts/s versus 13.5 Gpts/s for PT/Triton under random initialization, while zero-initialized runs reach up to 35.8 Gpts/s due to hardware memory compression. Using roofline and memory-hierarchy profiling, we show that -as expected- performance is predominantly bandwidth-limited and that compiler-managed L1/LSC caching can effectively replace programmer-managed shared-memory staging for this stencil class. Overall, the results position TinyTC as the performance-oriented path on Intel hardware and PyTorch/Triton as a strong portability/productivity baseline for cross-vendor HPC stencil development.
Read the full abstract

Algebraic Geometry

When Finite Free Curves Split

We characterize equality in the finite free Stam and entropy-power inequalities, proving that Hermite polynomials are the unique extremizers among simple real-rooted inputs, up to independent translations and scalings. The proof turns this classification into a rigidity problem for projective plane curves. Using hyperbolicity and the Helton-Vinnikov theorem, we express the Jacobian defect as an off-diagonal squared norm in a definite symmetric pencil. Together with score transport, this yields a matrix proof of Stam for all real-rooted inputs. For each simple base pair, the directions in which the defect vanishes are the independent translations and common dilation, forming a three-dimensional subspace in every degree whose curves split into $n$ projective lines. At a collision, the leading configurations are again finite free convolutions of normalized derivatives of the velocity polynomials of the colliding input clusters. Combined with incidence counting, this local formula shows that at most one real fiber is singular, every real singularity is an ordinary totally real multiple point, and the ordered collision multiplicities determine the real normalization covering. For $n\geq3$, every non-split curve has at least $2n-2$ non-real projective discriminant zeros, counted with multiplicity, with equality attained by irreducible curves of geometric genus zero through every simple input pair. The leading Fisher-information coefficient is determined by the colliding tangent configurations, while the finite entropy term retains the gaps between clusters. For $n\geq3$, maximal logarithmic entropy divergence along the optimally weighted score direction is equivalent to Stam equality.
Read the full abstract

Audio and Speech Processing

AVSRBench: A Multi-Condition AVSR Benchmark

While AVSR has achieved sub-1% word error rates on the standard LRS3 benchmark, its reliance on broadcast speech obscures whether this reflects true generalization or just domain adaptation. To investigate this gap, we evaluate three AVSR architectures across six conditions: controlled broadcast speech, fixed-grammar utterances, hyper-articulated Lombard speech, read speech from professional lipspeakers and non-professional speakers, and spontaneous multi-party video conversations. We find that visual-only performance deteriorates rapidly beyond broadcast domains, and audio-video fusion mainly benefits Lombard speech environments. Visual understanding degrades sharply at 90{\deg} profile views, with multimodal systems relying largely on acoustic fallback. Additionally, speaker articulation proves more critical than minor camera shifts, and LLM-based architectures suffer from poor out-of-domain generalization. Our work highlights a significant generalization gap in current AVSR research. To address this, we also introduce RoomReader-AV as a new benchmark for AVSR and release a unified data preprocessing pipeline to make comprehensive multi-condition evaluation accessible.
Read the full abstract

General Relativity and Quantum Cosmology

Universal Secular External Leg Corrections for Gauge Independent Scalar Self-Mass on de Sitter

This work concerns a procedure for removing gauge dependence from graviton corrections to the effective field equations for a massless, minimally coupled scalar on de Sitter background. The procedure involves combining diagrams in the same way as the scattering amplitude for the $t$-channel exchange of the massless scalar between two massive particles, but without taking the asymptotic limits which are problematic in cosmology. Implementing this at 1-loop on flat space background requires five classes of diagrams, in addition to the naive exchange, and the combination of those diagrams does eliminate dependence on the graviton gauge fixing functional. When those same diagrams are generalized to de Sitter background, their combination cancels the most important (“nonlocal”) secular gauge dependence, but it leaves (“local”) secular gauge dependence that is associated with external legs. Local secular gauge dependence can be removed by adding two additional classes of diagrams, and making secular corrections to the external wavefunctions. The purpose of this paper is to include these additional corrections on de Sitter background in order to derive fully gauge independent effective field equations at 1-loop order.
Read the full abstract

All authors D. Glavan, S. P. Miao, T. Prokopec, R. P. Woodard