Analysis of PDEs
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Sunday 30 August
In this paper, we consider energy identity for intrinsic stationary biharmonic maps into homogeneous spaces in supercritical dimensions, extending the corresponding result of Hornung-Moser [Anal. PDE. 2012] in critical dimension. The proof follows a similar strategy as that of Lin-Rivière [Duke Math. J. 2002]. A key ingredient is a conservation law for intrinsic biharmonic maps into homogeneous spaces, which allow us to derive higher regularity of the map.
Quantum Physics
Asymptotic Entanglement across Dynamical Regimes of a Dissipative Bosonic Dimer
Dynamical instability in parametrically driven bosonic systems is generally associated with diverging occupations and the absence of a stationary state, but its implications for long-time entanglement remain unclear. We investigate this question in a dissipative bosonic dimer with beam-splitter and two-mode-squeezing interactions, local loss, and a tunable common bath. The non-Hermitian dynamical spectrum governing the Gaussian moments partitions the parameter space into distinct regimes and provides a unified description of population and entanglement dynamics across both stable and unstable regions. We find that dynamical instability does not imply unbounded entanglement growth: although the bosonic population diverges, the logarithmic negativity remains bounded and approaches a finite asymptotic value. We further characterize the finite asymptotic entanglement and show that asymmetric local dissipation shifts both the stability boundary and the threshold for nonzero asymptotic entanglement. For identical local dissipation, the presence of a common bath leaves the dynamical-region boundaries unchanged while qualitatively reshaping the long-time entanglement landscape, producing broad enhancement regions within the stable regimes and a narrow enhancement region within the unstable regime. In the common-bath-only limit, this enhancement is attributed to an exact dark-mode (bound-state-in-the-continuum, BIC) protection mechanism, which reduces to a quasi-BIC remnant under independent dissipation. Our results establish the non-Hermitian dynamical spectrum as a framework for connecting stability, nonequilibrium dynamics, and asymptotic entanglement in open quadratic bosonic systems.
Read the full abstract
Analysis of PDEs
Plasmon poles for the linearized Coulomb–Hartree–Fock equation
We study the time-dependent Hartree–Fock equation in $\R^3$ with physical Coulomb interactions in both direct and exchange channels, with exchange coupling $\eta$, linearized about homogeneous equilibria $\gamma_f=g(-i\nabla)$ having compact momentum support. Nguyen and You showed that, without exchange, two purely imaginary plasmon poles $\pm i\tau_0(|k|)$ exist below a survival threshold $\kappa_0>0$. Existing nonlinear Hartree–Fock theory requires a short-range direct interaction and a smooth small exchange kernel, excluding the present Coulomb setting. On every compact band $0<k_-\le|k|\le k_+<\kappa_0$, we prove unconditionally for Coulomb exchange that, for small $|\eta|$, the poles persist, remain simple and purely imaginary, and depend real-analytically on $\eta$. Their first-order shift $\tau_X$ splits into explicit static self-energy and dynamic vertex corrections. Each pole yields an exact undamped mode with a closed-form unit-density eigenfunction, embedded in the fiber generator's essential spectrum. The causal density Green function splits into an explicit undamped sine wave and a remainder whose Laplace transform is holomorphic near the poles, while the plasmon channel satisfies uniform $t^{-3/2}$ dispersive decay and endpoint Strichartz estimates. A first-order Ward-type cancellation gives $|\tau_X(r)|\le Cr^2$ down to $r=0$; hence exchange leaves the plasma gap unchanged to first order, and we obtain a closed formula for the stiffness correction. Numerics for a $C^9$ smoothed Fermi ball support the analysis: self-energy and vertex terms cancel within $1.2\%$ on the computed range, and exchange softens the dispersion most strongly near $0.7\,\kappa_0$.
Read the full abstract
Complex Variables
Uniformizing non-proper Gromov Hyperbolic Spaces
In this paper, we extend a large part of the uniformization theory of Bonk-Heinonen-Koskela [Asterisque 2001] to length spaces that are not necessarily proper or geodesic. Among other things, we show that there is a one-to-one correspondence between the quasiisometry classes of complete roughly starlike Gromov hyperbolic spaces and the quasisimilarity classes of bounded uniform spaces, which provides an affirmative solution to an open question of Bonk-Heinonen-Koskela. Our approach relies crucially on the work of Väisälä [Expo. Math. 2005], who investigated in depth Gromov hyperbolic spaces that are not necessarily proper or geodesic. One key new ingredient is to use the so-called (quasihyperbolic) $(c,\mu)$-quasigeodesic as a suitable substitute for quasihyperbolic geodesic.
Read the full abstract
Statistical Finance
Who Are a Firm's Economic Peers? Mapping a Latent Information Field with Language Models
Identifying a firm's economic peers matters for benchmarking, risk measurement, and understanding return dependence, yet a diversified firm may have no single natural peer. We use language-model representations of corporate news to identify the combination of firms that jointly represents each firm's information environment. Each firm is represented by a distribution of article embeddings, and target-anchored Wasserstein barycentric reconstruction produces a directed map of economic-peer weights. We use this map as the interaction structure in a spatial factor model and ask whether return dependence follows peer relations measured from news without using returns. We construct peer weights from 2018-2022 news for 100 large-cap US traded firms and evaluate their financial relevance in subsequent returns. Relative to a matched model with the Fama-French five factors and momentum, the information field firms' unexpected returns move together after common factor exposures are accounted for. Joint reconstruction contributes substantially more than pairwise Wasserstein proximity or equal weighting of the selected peers, while centroid reconstruction retains most of the gain. Explicit news co-mentions provide complementary information. Language models thus make the interaction structure in spatial asset pricing an object of measurement: economically interpretable peer relations can be constructed from text before their financial relevance is assessed in returns.
Read the full abstract
Instrumentation and Methods for Astrophysics
Microchannel plate detector development for ultraviolet astronomy
Observational data from the UV wavelength range is crucial to solve several astrophysical puzzles. Consequently, there are a number of upcoming UV missions from CubeSats up to the future flagship Habitable Worlds Observatory. Besides novel instrument concepts and improved coatings, advanced detectors are key for the success of these missions. The microchannel plate (MCP) detector technology offers a unique asset in the UV: the combination of single-photon counting and visible-blindness. The UV hardware group at the Institut für Astronomie und Astrophysik Tübingen (IAAT) develops a versatile MCP detector system that addresses the complete UV band. While a sealed-tube design is limited to wavelengths above 118\,nm, an open-face variant also covers the whole far- and extreme-UV. Both use the same readout: a coplanar cross-strip anode and FPGA-based electronics. In this contribution, we report on the status of the detector development, present the latest characterization results, and give an outlook on the mission prospects.
Read the full abstract
Machine Learning
A Target-Centric Survey of Quantization-Aware Training
The rapid development of LLMs incurs prohibitive memory footprints and intensive computational demands. Quantization-Aware Training (QAT) techniques have emerged as a promising solution to address these challenges by explicitly simulating quantization effects during model training, yielding low-bit models that achieve accuracy comparable to their full-precision counterparts. In this work, we provide a target-centric survey of QAT, aimed at clarifying both its theoretical foundations and its evolving implementation landscape. We systematically review existing QAT methods through a target-centric taxonomy and synthesize cross-target differences in error characteristics, numerical formats, and strategy transferability. We further summarize QAT evaluation paradigms and discuss challenges in optimization and deployment, outlining potential directions for future research.
Read the full abstract
High Energy Physics (Phenomenology)
Probability-based Estimates of the Uncalculated N$^5$LO QCD Contribution to the Hadronic $W$-Boson Decay Width
The perturbative QCD corrections to the hadronic decay width of the $W$ boson are currently known up to next-to-next-to-next-to-next-to-leading order ($\mathrm{N^4LO}$), whereas the exact $\mathrm{N^5LO}$ correction remains unavailable owing to its formidable computational complexity. In this work, we estimate the $\mathrm{N^5LO}$ contributions by adopting Bayesian analysis (BA). Before performing the estimation, the Principle of Maximum Conformality (PMC) is employed to improve the precision of the initial scale-dependent perturbative series. Through recursive application of the renormalization group equation, non-conformal terms are absorbed into the strong running coupling, yielding a scheme-independent, scale-invariant perturbative series with improved convergence. The PMC procedure determines an effective coupling $\alpha_s(Q_*)$, with the PMC scale fixed as $Q_* = 100.102~\mathrm{GeV}$ at next-to-next-to-leading logarithmic accuracy. Based on the improved and more precise series, the $95.5\%$ BA credible interval yields an uncertainty of $\Delta\delta_\mathrm{QCD}|_{\mathrm{PMC, BA}}^{\mathrm{N^5LO}} = \pm 1.0\times 10^{-5}$. The resulting hadronic branching ratio is $\mathcal{B}(W\to \text{hadrons})|_\mathrm{PMC} = (65.84\pm 1.54)\%$, which is consistent with experimental data within reasonable errors.
Read the full abstract
Materials Science
Evaluating a 4B open-weights local LLM for agentic DFT workflows: a literature reproducibility audit
Agentic workflows in materials science relying on hosted commercial models face severe reproducibility, economic, and data-privacy constraints. To explore fully local agentic science, this work evaluates an open-weights Qwen3:4B model executing an autonomous scientific pipeline across varying hardware constraints. Applied to pentagonal two-dimensional materials, the system extracts parameters from unstructured text, translates them into density functional theory (DFT) inputs, and drives simulations to convergence under a strict neurosymbolic architecture where agents propose and deterministic code disposes. The workflow is guarded by verbatim text grounding and multi-pass inference unions to counteract hardware-induced structural collapse. Evaluated against 201 expert judgements, the extractor achieves 95.7% precision (95% CI 90.3-98.1%) and 67.3% recall (59.8-74.0%), ensuring extracted parameters are strictly factual. However, precision identifying absent parameters does not exceed 47.0%, establishing that the measured omission rate constitutes a loose upper bound on true literature incompleteness. Across three hardware configurations, complete GPU residency governs extraction quality more fundamentally than weight or cache precision, raising Matthews correlation from 0.414 to 0.530 at fixed quantisation and to 0.560 with an unquantised cache. A corpus-scale audit indicates only 19 (33.3%) of the 57 studies are reproducible in principle, reporting every method parameter needed to re-initialise the calculation. Driven to convergence, the workflow reproduces published lattice constants with a mean absolute relative error of 2.3% where the relaxed structure retains its prototype, establishing that lightweight open-weights models can reliably drive autonomous agentic workflows when bounded by deterministic code gates.
Read the full abstract
Analysis of PDEs
Critical Morrey Rigidity and Removable Singularities for Five-Dimensional Stationary Navier-Stokes Flows
We prove a critical Morrey rigidity theorem for the five-dimensional stationary Navier–Stokes equations. More precisely, every smooth solution on $\mathbb R^5\setminus\{0\}$ satisfying \[ \sup_{R>0}R^{-2}\int_{B_R}|u|^3\,dx<\infty \] is identically zero, up to an additive constant in the pressure. This replaces the pointwise Type-I control in the known higher-dimensional rigidity theory by a velocity-only, scale-invariant averaged condition that allows spatial concentration. The proof develops a weak head-pressure mechanism that does not rely on pointwise pressure estimates or classical normal traces. We reconstruct a canonical pressure from the velocity, derive a renormalized inequality for the positive head pressure, and introduce two monotone radial fluxes. Annular energy estimates, suitable-weak compactness, and blow-up and blow-down limits are then used to identify the endpoint fluxes and force rigidity. As an application, we obtain a removable-singularity criterion in dimension five: if a suitable weak solution is smooth away from one point and either its scale-invariant Dirichlet energy or its cubic velocity Morrey quantity remains bounded near that point, then the singularity is removable. Thus, within the isolated-singularity class, the smallness assumption in the classical stationary regularity criterion is replaced by boundedness. We also prove the corresponding velocity-only cubic Morrey rigidity theorem in dimension four by a different finite-energy argument.
Read the full abstract
Computer Vision
PhysVR: Vision-Language Model Guided Interference-aware Temporal Feature Refinement for Remote Physiological Measurement
Remote photoplethysmography (rPPG) enables contactless physiological measurement from facial videos, yet its subtle pulse-related variations are easily affected by illumination variation, head motion, facial blur, and region-of-interest instability. Existing methods mainly suppress interference during feature learning, while whether the learned temporal features remain affected by interference and how to further suppress such interference before rPPG estimation are rarely examined. To address this limitation, we propose PhysVR, a vision-language model guided interference-aware temporal feature refinement framework for rPPG estimation. Specifically, a physiological backbone produces global temporal features and a coarse rPPG prediction, from which signal-derived physiological reliability evidence is constructed from local temporal characteristics. In parallel, a frozen vision-language model processes sampled facial frames under an interference-oriented prompt, and an evidence head extracts visual interference evidence from the VLM output. Temporal cross-attention integrates the physiological and visual evidence with the global temporal features to construct interference-aware temporal context. Guided by this context, a shared temporal correction unit performs general refinement, while four interference-specific experts selectively suppress different interference through adaptive routing. The refined temporal features are then used for final rPPG estimation. Extensive experiments on five public benchmarks demonstrate that PhysVR consistently outperforms representative methods under both intra-dataset and cross-dataset evaluation protocols.
Read the full abstract
Computation and Language
ACTD: Anchor-Based Cross-Tokenizer Distillation with Residual Regularization
Knowledge distillation effectively transfers reasoning capabilities from large language models to lightweight student models. To enable knowledge transfer across disparate model families, researchers increasingly explore cross-tokenizer distillation. However, cross-tokenizer distillation remains challenging due to vocabulary and sequence misalignment, while approximate vocabulary alignment can introduce additional noise into distillation. To address these challenges, we propose Anchor-Based Cross-Tokenizer Distillation with Residual Regularization (ACTD). ACTD bridges structural heterogeneity through vocabulary and sequence alignment, while mitigating alignment noise via a novel anchor loss with residual regularization. We further extend this framework to a multi-teacher setting. Evaluated across five reasoning benchmarks with three distinct teacher models, ACTD achieves state-of-the-art performance. Moreover, its multi-teacher extension outperforms the strongest single-teacher and multi-teacher baselines, further demonstrating the robustness of our method.
Read the full abstract
Computational Engineering
OmniClimate-TC: Physics-Aware Visual Abstractions for Multimedia Reasoning over Tropical Cyclones
Meteorological reanalysis encodes extreme weather through continuous, physically constrained fields, posing a fundamental challenge for vision-language models (VLMs) whose perceptual assumptions are shaped by natural images. Tropical cyclones exemplify this mismatch: critical properties such as intensity extrema, asymmetry, spatial extent, and physical impacts arise from field-level organization rather than object-centric visual cues. Existing approaches address this gap through text alignment or annotation, treating the problem as multimodal supervision rather than representation design. We introduce Physics-Aware Visual Abstraction (PAVA), a plug-and-play physics-aware representation and annotation interface that maps physical reanalysis fields to visually identifiable and semantically grounded perceptual abstractions for supervision and evaluation in vision-language reasoning. Building on PAVA, we construct OmniClimate-TC, a benchmark for tropical cyclone analysis spanning five classes of reasoning and nine tasks, with 243,890 physically grounded instruction-tuning pairs. Using PAVA-aligned supervision, we adapt VLMs and provide evidence that this representation design improves reasoning over tropical cyclone hazard fields. Our results position OmniClimate-TC as a benchmark for multimedia reasoning over structured geophysical fields, and highlight representation design as a key ingredient for physically grounded reasoning in scientific media.
Read the full abstract
Functional Analysis
The Beurling density of the spectrum of self-similar measure generated by Hadamard triple
Let $D\subset\Bbb Z$ with cardinality $q\ge 2$, and let $b\in \Bbb Z$ with $q<b$, and let $\mu:=\mu_{b,D}$ be the associated self-similar measure. It is well known that if there exists $L\subset \Bbb Z$ such that $(b,D,L)$ be a Hadamard triple, then the Beurling dimension of the spectrum of $\mu$ exhibits an intermediate structural property. In this paper, we establish a stronger result that both Beurling dimension and Beurling density of the spectra of $\mu$ can achieve full flexibility simultaneously. More precisely, for any $t\in(0, \frac{\log q}{\log b})$ and $s\in [0,\infty]$, there exists a spectrum $\Lambda:=\Lambda_{t,s}$ of $\mu$ such that $$\dim_{Be}(\Lambda)=t,\quad D_t^+ (\Lambda)=s.$$ Here, $\dim_{Be}$ and $D_t^+$ denote the Beurling dimension and the $t$-Beurling density, respectively. We further prove that the set of such spectrum whose Beurling dimension and Beurling density are equal to any fixed $t$ and $s$ has the cardinality of the continuum. \par This work generalizes a previous result of Lu \cite{Lu}, answers an open question raised by Dai, Fu and He \cite[Conjecture 5.3]{DaiFuHe}, and sheds new light on the fine structural properties of spectra for singularly continuous spectral measures.
Read the full abstract
History and Overview of Mathematics
On the Babylonian Division of Trapezoids
We discuss the Babylonian division of trapezoids and the construction of such trapezoids with rational sides and transversals from Pythagorean triples.
Physics and Society
ButterMamba: Butterworth-Enhanced Spatial-Temporal Mamba for Efficient Traffic Flow Prediction
Accurate traffic flow prediction is fundamental to intelligent transportation systems, playing a pivotal role in urban mobility optimization and smart city development. While Graph Neural Networks (GNNs) integrated with time series forecasting have emerged as promising solutions, two critical limitations persist: (1) the quadratic complexity of attention-based architectures hinders real-time deployment in large-scale networks, and (2) high-frequency noise in sensor data significantly degrades prediction reliability. These challenges are particularly acute in metropolitan scenarios where both computational efficiency and noise robustness are paramount. To address these limitations, we introduce ButterMamba, a novel and efficient framework based on State Space Models (SSMs). ButterMamba consists of two key components: (1) a Butterworth Spectral Filtering module that preprocesses the data by removing high-frequency noise, allowing the model to focus on significant underlying trends, and (2) a Spatial-Temporal State Mixer that uses a parallel Mamba architecture to efficiently capture both long-range temporal dependencies and complex spatial correlations across the road network. By decoupling noise filtering from spatial-temporal modeling, ButterMamba achieves superior predictive accuracy with linear computational complexity. Extensive experiments on three public datasets demonstrate that ButterMamba not only outperforms existing state-of-the-art models in terms of prediction accuracy but also considerably reduces training time and memory usage.
Read the full abstract
Probability
Hypocoercivity of Tempered Bouncy Particle Samplers for Heavy-Tailed Targets
We introduce the Tempered Bouncy Particle Sampler, a state-dependently tempered generalisation of the Bouncy Particle Sampler. Adapting the hypocoercivity framework of Dolbeault, Mouhot and Schmeiser (2010), we establish explicit, non-asymptotic exponential convergence bounds for a broad class of heavy-tailed targets that need not be log-concave or radially symmetric, including distributions with polynomial tails. The convergence rate is linked to the spectral gap of a tempered Langevin diffusion with the same invariant distribution, making a weighted Poincaré inequality the key assumption. We give admissible tempering choices for polynomial and stretched-exponential tails and derive mixing-time bounds with polynomial dependence on the dimension.
Read the full abstract
Distributed Computing
A Smallest-Need-First Job Scheduling Framework with Adaptive Optimization of Idle Node Counts for Energy-Efficient HPC Systems
Power-state management in high-performance computing (HPC) clusters must reduce idle energy without excessive wake-up delays for rigid parallel jobs. This paper presents SNF-ICON, an event-driven controller combining smallest-need-first (SNF) gang scheduling, predictive wake timing, and adaptive warm-spare control. At each scheduler invocation, recent interarrival and completed-service samples are screened for sufficiency, exponential-like variability, low lag-one autocorrelation, and acceptable Kolmogorov-Smirnov distance. Rejected or data-sparse windows use SNF+IPM (Intelligent Power Manager), whereas accepted windows activate release prediction and an exponential next-event model. Warm-spare optimization is applied only when queue, event, and arrival-recency conditions permit, balancing estimated waiting and non-compute energy over a timeout-capped horizon. We evaluate four DAS2 trace segments and a generated Markovian workload on AOBA-derived 64-node models, plus SDSC Blue on an AOBA-derived 1152-node model. SNF-ICON is compared with SNF+IPM and First Come First Served (FCFS) + backfilling with IPM. It reduces average waiting time relative to the FCFS-based baseline in all six cases and remains close to at least one heuristic energy baseline in five. The generated workload spends substantial time in ICON mode, whereas DAS2 workloads operate mainly in fallback. Furthermore, cross-platform results show strong dependence on node-transition and power models. Thus, no single policy or parameter set works best in every case.
Read the full abstract
Combinatorics
Triangulated polygons and Y-frieze patterns
In the spirit of Conway and Coxeter, we classify all $\mathbf{Y}$-frieze patterns of type $A_n$. In particular, we settle a conjecture made by de Saint Germain that all such $\mathbf{Y}$-frieze patterns arise from Conway-Coxeter frieze patterns. Moreover, our approach naturally leads to the enumeration of these $\mathbf{Y}$-frieze patterns in terms of Fuss-Catalan numbers.
Read the full abstract
Numerical Analysis
Constraint Preserving AFD-WENO Schemes for Relativistic Hydrodynamics with General Equations of State
We develop a high-order physical-constraint-preserving (PCP) alternative finite difference weighted essentially non-oscillatory (AFD-WENO) scheme for the special relativistic hydrodynamics equations with general equations of state. The proposed scheme comprises two key limiters: a state limiter, which acts after the WENO state interpolation step, and a flux limiter, which acts on the final high-order fluxes. The state limiter ensures that the interpolated states are physically admissible, while the flux limiter ensures that the numerical fluxes are physically admissible. The resulting scheme is rigorously proved to satisfy the physical constraints. Incorporating multiple WENO interpolation techniques, including an improved adaptive-order formulation (WENO-AOI), the method is validated through extensive one- and two-dimensional numerical benchmarks with various equations of state. The numerical results demonstrate high-order accuracy, sharp resolution of discontinuities, and robust stability in extreme relativistic regimes.
Read the full abstract
Plasma Physics
Generation of Isolated Collimated Polarized $\gamma$-ray Beams via Spatiotemporal Optical Vortex Modulation
Attosecond, collimated, bright, polarized $\gamma$-ray sources are in high demand across nuclear physics, astrophysics, and high-energy physics. However, realizing attosecond duration, high collimation, high brilliance, and high polarization simultaneously within a single isolated source remains an outstanding challenge, owing to the inherent trade-offs between beam trapping and radiative dynamics. Here, we propose a novel scheme to generate an isolated, collimated, high-brilliance, polarized attosecond $\gamma$-ray beam from conventional solid foils irradiated by a linearly polarized spatiotemporal optical vortex (STOV) laser pulse accessible in Lab. Three-dimensional spin-resolved particle-in-cell simulations reveal that this relativistic-intensity STOV pulse can trap and accelerate electrons at its spatiotemporal singularity, producing a compact isolated electron bunch. This electron bunch subsequently undergoes head-on collision with the reflected laser pulse, which generates isolated $\gamma$-ray beams through nonlinear Compton scattering. With a peak intensity of $7\times10^{21}$ W/cm$^2$, we observe an isolated collimated ($\sim1.5^{\circ}$) $\gamma$-ray beam with an average linear polarization of $>60\%$ and a duration of $\sim$500 attoseconds. This approach is feasible with current or upcoming laser facilities and robust against variations in laser and target parameters, highlighting the capability of spatiotemporal structured light field modulation to address outstanding problems in plasma physics.
Read the full abstract
Information Retrieval
ICEGR: An Intent-Coherent End-to-End Generative Retrieval Framework for E-commerce Search
Generative Retrieval (GR) is promising for e-commerce search, yet existing methods struggle to maintain query-intent consistency throughout the training pipeline. First, semantic ID (SID) construction based on static product information limits the ability of SIDs to encode product-intent associations. Second, although supervised fine-tuning (SFT) learns product-SID mappings across the catalog, low-exposure products still lack real query-intent supervision because query-to-SID training relies solely on online logs, resulting in poor retrieval performance for these products. Third, business-oriented preference optimization may favor popular or high-value products over those that best match the query intent, weakening query-product relevance. To address these issues, we propose ICEGR, an Intent-Coherent End-to-End Generative Retrieval Framework for E-commerce Search that integrates query intent consistently throughout the GR training pipeline. ICEGR comprises three components: (1) Intent-Aware SID Construction incorporates query-intent signals into SID construction, enabling SIDs to capture search intent beyond static product information; (2) Synthetic Query-Enhanced Unified SFT unifies multiple SFT tasks under the query-to-SID objective and augments sparse supervision from online logs with synthetic queries, providing complementary query-intent supervision for low-exposure products; and (3) Relevance-Calibrated Preference Optimization integrates query-product relevance and business signals into a margin-adaptive preference objective, preserving query intent while enabling business preference learning. Offline results show that ICEGR improves Recall@20 by 21.7% and NDCG@20 by 26.6% over the baseline. Deployed as an end-to-end generative retrieval pathway in Baidu E-commerce Search, ICEGR achieves relative improvements of 3.52% in CTR, 15.96% in order volume, and 7.53% in GMV in an A/B test.
Read the full abstract
Differential Geometry
Remarks on the Complex Structures on $\mathbb P^3$ and $S^2\times S^4$
Assuming the validity of the recently proposed “A compact complex threefold fibred by tori over the projective line, and the six-sphere”, we construct an exotic complex structure on $\mathbb P^{3}$, distinct from the point-blowup structures of Huckleberry, Kebekus, and Peternell. Then we perform an Atiyah flop to produce a complex structure on the standard smooth manifold $S^{2}\times S^{4}$.
Read the full abstract
Systems and Control
On the Number of Observation Nodes in Recurrent Neural Networks with Linear Threshold and ReLU Functions
This paper investigates how node update rules and admissible state domains affect the minimum number of observation nodes required for global finite-horizon observability in recurrent neural networks with linear-threshold and ReLU update functions. Over a common binary state domain, we construct a class of $K$-linear-threshold ($K$-LT) networks whose initial states can be uniquely reconstructed from the finite output trajectory of a single observation node. We further establish a dynamical equivalence between binary-valued $K$-ReLU networks and $K$-AND Boolean networks, which transfers existing class-level observation-node bounds to binary-valued $K$-ReLU networks. For nonnegative-valued ReLU networks, the observability problem reduces to the classical linear-system setting whenever the relevant pre-activations remain nonnegative. For general real-valued ReLU networks, we prove that global finite-horizon observability requires at least $n/2$ observation nodes when no restriction is imposed on the number of state variables involved in each node update. This lower bound is tight when $n=2K$, for which we construct a $K$-ReLU network observable from exactly $K$ nodes. These results show that both update rules and state domains fundamentally affect extremal observation requirements: temporal evolution can concentrate finite-state information into a single measured trajectory, whereas activation-induced rank loss creates an intrinsic sensor lower bound in continuous-state ReLU networks.
Read the full abstract
Solar and Stellar Astrophysics
Origin of small-scale evaporation flows deep in the chromosphere during a solar flare
Flares are caused by an abrupt release of magnetic energy in the solar atmosphere. Plasma heated to well over 10 MK filling the post-flare corona originates from a rapid heating and ablation of the cooler chromospheric material. This chromospheric evaporation is thought to be facilitated primarily by nonthermal electrons impinging on to the lower atmosphere. Questions on when and where in the chromosphere these upflows originate, however, are not fully resolved. Here we report on unprecedented high-resolution observations of an M-class flare recorded by the Sunrise Ultraviolet Spectropolarimeter and Imager on board the balloon-borne SUNRISE observatory, that reveal highly structured upflows on spatial scales of ~100 km originating deep in the chromosphere. The flows even precede the onset of nonthermal electrons by about 10 minutes and last through the impulsive phase of the flare. Our observations shed new light on the lower atmospheric heating and mass circulation in flares that are challenging to reconcile with the standard solar flare model.
Read the full abstract
Instrumentation and Detectors
Toward Ionization Cluster Size Measurements with a Compact Nanodosimeter
Nanodosimetry aims to provide measurable quantities related to the nanoscopic particle track structure, which determines the biological effectiveness of radiation. While simulated nanodosimetry has already demonstrated its potential for radiation treatment planning, the experimental realization of practical nanodosimetric detectors is still in its early stages. In this work, a nanodosimetric prototype operated with low-pressure gas was developed to count ionizations in a nanometer-equivalent sensitive gas volume. Its performance was evaluated experimentally with alpha beams from an Am-241 source in 1 mbar propane gas. The results support the further development of this compact nanodosimeter class, with potential applications in particle therapy, radiation protection, and space radiation dosimetry.
Read the full abstract
Machine Learning
Reward-guided Fine-Tuning of One-Step Generative Models via Wasserstein Gradient Flow
To mitigate the time complexity of generative models, one-step generative models have recently emerged through direct mapping from noise to data in a single forward pass. However, the reward-guided fine-tuning method of one-step generative models remains largely unexplored. To address this, we consider one-step generators from an optimal transport view, investigating Wasserstein Gradient Flow (WGF) for modeling smooth and controlled distributional evolution in probability space. We then propose a novel reward-guided fine-tuning of a one-step generative model via WGF. We derive a practical training method that requires no reward gradients, thereby handling both non-differentiable and differentiable rewards. Moreover, our method provides smooth and stable reward-guided distributional updates while mitigating reward hacking and mode collapse. Experiments on 2D synthetic data, CIFAR-10, and ImageNet 256$\times$256 with diverse rewards, including JPEG (in)compressibility, class probability, Black-and-White and CLIP alignment, show that our method achieves better reward alignment compared to baselines.
Read the full abstract
Artificial Intelligence
Cascade Failure Attribution for Multi-Agent Systems
Large language model (LLM)-based agents have shown strong potential in solving complex tasks through multi-step reasoning, yet they remain vulnerable to execution failures. Accurate failure attribution is therefore critical for improving agent reliability. Existing topology- and spectrum-based methods exploit trajectory structures but often overlook fine-grained semantics, while LLM-based attribution methods capture semantic cues but suffer from long-context degradation over lengthy trajectories. To address these challenges, we propose DuoTrace, a plug-and-play detection filter for LLM-based failure attribution. DuoTrace follows a detect-before-attribute paradigm: it first detects anomalous executions and then supplies focused trajectory evidence to downstream LLM-based attribution methods. For effective VAE-based anomaly detection on agent trajectories, DuoTrace integrates dual-view semantic-structural node representations, a Tree-LSTM-based trajectory encoder, and data augmentation through outcome-verified LLM replay and root-to-node path extraction to handle heterogeneous nodes, hierarchical execution structures, and limited normal execution data. Experiments with six LLM-based attribution baselines show that DuoTrace improves agent-level and step-level attribution accuracy by 11.2% and 7.7%,respectively.
Read the full abstract
Mathematical Physics
A Morse-Family Integrator for Hamilton–Jacobi Dynamics Across Caustics
We develop a geometric framework for implicit discrete Hamiltonian systems based on discrete Morse families, Lagrangian relations, and discrete analogues of Tulczyjew's triple. The main idea is to regard the Lagrangian submanifold defining the discrete dynamics, rather than an explicit symplectic evolution map, as the fundamental geometric object. This viewpoint naturally accommodates implicit, constrained, and degenerate discrete systems. Within this framework, we formulate a Type–II discrete Hamilton–Jacobi theory in terms of the propagation of Lagrangian submanifolds between consecutive discrete steps. When these submanifolds are locally represented by exact one-forms $dW_k$ and $dW_{k+1}$, the resulting equations provide a discrete Hamilton–Jacobi relation between consecutive generating functions. More generally, when the projection onto configuration space becomes singular and a single-valued generating function ceases to exist, we show that the evolution can be described by the composition of Type–II discrete dynamics with Morse families. This yields a generating family for the propagated Lagrangian submanifold without requiring the dynamics to be represented as a graph. As an application, we consider the propagation of optical wavefronts through fold caustics. A Type–II discrete Hamiltonian yields a symplectic ray integrator, while a Morse family represents the multivalued wavefront near the caustic. Their composition provides a discrete propagation rule for the complete Lagrangian manifold across the singularity. In this way, the same geometric construction simultaneously provides a discrete Hamiltonian integrator and a regular representation of multivalued Hamilton–Jacobi solutions.
Read the full abstract
Computer Vision
Conducting Stylistic Analysis of Paintings through an Art-History Agent
Attributing an artwork to an artist has traditionally relied on detailed visual observations and descriptions, known as stylistic analysis in art history. By contrast, current artificial intelligence (AI) models used in the field offer only unexplained probabilistic classifications. To bridge this methodological gap, we present an AI framework that automates stylistic analysis of paintings, providing a foundation for enhancing evidence collection, discovery, and verification. By training a vision transformer (ViT) on a large corpus of paintings with metadata, our system encodes this art history-specific data as embeddings. These representations are factorized via sparse dictionary learning into a shared set of features that recur across the training set. A large language model (LLM) then interprets each feature by retrieving associated artworks and their accompanying curator-written texts, and synthesizes them into descriptions that reflect their stylistic attributes. Finally, an autonomous coordinator LLM applies a reasoning-and-action (ReAct) framework to weight, test, and refine these features into cohesive descriptions of an artwork, or comparisons of artworks. This approach converts detailed visual features into descriptive terms, addressing a key challenge in art history. It thus connects the use of images as data with the semantic concerns of humanists, establishing vision-based computational art history as an area for future growth.