Number Theory
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Tuesday 15 September
We obtain a Furstenberg-Sárközy-type result for sets $A\subset [N]$ whose difference set $A-A$ does not contain the sum of $s$-many $k$-th powers of positive integers, with $k>1$ odd and $s>0$ even. Namely, we prove that such sets must satisfy a power-saving bound $|A| \, \ll \, N^{1-\frac1k\min\{s \, \sigma_k, \, 1/2\}+\epsilon}$ for any fixed $\epsilon>0$, where $\sigma_k >0 $ is any admissible saving in a classical one-variable Weyl estimate. In particular, we can take $\sigma_k=\max\left\{{2^{1-k}}, \, \frac{1}{k(k-1)}\right\}$ using the classical theory and the best currently available bounds for classical Weyl sums. A greedy construction produces a set $A\subset[N]$ with $|A|\gg N^{1-s/k}$ for which $A-A$ contains no sum of $s$-many positive $k$-th powers, so our power-saving bound is of the correct shape.
Computer Vision
FRPSS: Feature Rearrangement in Pre-Shape Space for Single-Image Generation
Generative models trained on a single image often struggle to balance global structural integrity and local diversity. Existing single-image generation methods commonly rely on random noise to drive the generation process and lack explicit global structural constraints, making the generated results prone to spatial structural misalignment when structural variations occur. To address the issue, Feature Rearrangement in Pre-Shape Space for Single-Image Generation (FRPSS) is proposed in this paper. The core of FRPSS is the Manifold Structural Rearrangement with Feature Augmentation on Geodesic Surface (MSR-FAGS) module. MSR-FAGS replaces the randomly initialized features of the low-scale generator with rearranged Pre-Shape features and uses the features to guide image generation at subsequent scales, thereby reducing the risk of structural misalignment. To support downstream tasks such as stylization, a Scale-adaptive Sliding-window Patch Extraction (SSPE) strategy is further designed, and a directional Contrastive Language-Image Pre-training supervision module with SSPE (CLIP-SSPE) is constructed. Qualitative and quantitative experiments demonstrate that FRPSS achieves the best Single Image Fréchet Inception Distance (SIFID) scores on all three datasets while maintaining competitive Learned Perceptual Image Patch Similarity (LPIPS). Further qualitative experiments verify the effectiveness of FRPSS across multiple downstream tasks with the CLIP-SSPE module.
Read the full abstract
Quantum Gases
Weak ferromagnetism in the square-lattice Heisenberg $J_1$-$J_2$-$J_2^\prime$ model with easy-axis single-ion anisotropy
Motivated by recent experimental realizations of the Heisenberg model and two-dimensional altermagnetism using ultracold atoms in optical lattices, we systematically investigate the magnetic properties of the square-lattice Heisenberg $J_1$-$J_2$-$J_2^\prime$ model in the presence of easy-axis single-ion anisotropy $D$. Here, $J_1 (>0)$ is the nearest-neighbor antiferromagnetic exchange parameter, while $J_2$ and $J_2^\prime$ are two distinct next-nearest-neighbor superexchange parameters that alternate on the lattice. We perform numerical calculations of the sublattice magnetization, net magnetization, and critical temperature. Interestingly, the net magnetization is not identically zero at finite temperatures; its absolute value first increases and then decreases with increasing temperature, revealing a temperature-dependent weak ferromagnetism. We find that $J_2\not=J_2^\prime$ is the primary factor responsible for weak ferromagnetism at finite temperatures, while $D$ provides the necessary background for two-dimensional long-range magnetic ordering. Moreover, for a fixed $J_2+J_2^\prime$, we observe that a larger $|J_2-J_2^\prime|$ leads to a greater maximum net magnetization and critical temperature, as well as a broader temperature range for weak ferromagnetism. The difference between $J_2$ and $J_2^\prime$ originates from the distinct superexchange pathways between magnetic atoms, with these pathways mediated by two different types of nonmagnetic atoms. This mechanism, which differs from conventional ones such as the Dzyaloshinskii-Moriya interaction, may be detectable in real materials.
Read the full abstract
Artificial Intelligence
A Framework for Generating Valid Context-Specific Benchmarks through Expert Guidance
This paper presents an end-to-end approach for generating context-specific large language model (LLM) benchmark datasets by combining expert input with synthetic data generation. Existing benchmark construction methods often trade off validity and scalability: datasets designed with domain experts can produce high-quality evaluations but are slow and costly to create, while synthetically generating data may scale efficiently but often results in unrealistic, redundant, or out-of-scope examples. To address this gap, we introduce a schema eliciting key information about the goals, scope, and context of an evaluation task, and use this information to guide synthetic data generation. We further define four criteria grounded in measurement validity for assessing dataset quality: coverage, diversity, content realism, and stylistic realism. Using these criteria, we show how expert-informed scaffolds can guide synthetic data generation toward more valid benchmarks. Through quantitative evaluations and a real-world case study with domain experts, we demonstrate that our approach improves benchmark data quality over existing methods while preserving validity. We additionally analyze how different types of schema information affect different dataset quality criteria, and provide practical guidance on which information to prioritize collecting under resource constraints.
Read the full abstract
Computer Vision
FLAT: Resampling Image and Text into 1D Flexible-Length Aligned Transmodal Tokens for Retrieval and Generation
Traditional multimodal representation learning and generation are two stages: a contrastive or self-supervised visual encoder is trained first, followed by a separate downstream generative model. This setup bottlenecks generative performance behind frozen embeddings. To bridge this gap, we revisit joint multimodal representation learning and generation to produce linearly interpolatable embeddings that are directly consumable by generative decoders. We present FLAT (Flexible-Length Aligned Transmodal representations), a representation pre-training framework that jointly optimizes a shared multimodal encoder alongside downstream text-to-image (T2I) and image-to-text (I2T) decoders. By combining contrastive alignment with bidirectional cross-modal generative objectives, FLAT ensures its representations function as both discriminative semantic descriptors and generative conditions. Architecturally, FLAT maps visual and textual inputs into a unified continuous 1D sequence space, applying nested dropout over prefix-K tokens to enable dynamic output lengths. A single pre-training stage allows FLAT to perform cross-modal retrieval and generation across variable prefix K, achieving a T2I GenEval score of 71.1. Task-specific fine-tuning aligns model performance with state-of-the-art baselines: 83.1 GenEval on T2I generation; 40.5 BLEU-4 and 138.6 CIDEr on MS-COCO image captioning; and Recall@5 scores of 86.8 (I2T) / 75.8 (T2I) on MS-COCO alongside 98.3 (I2T) / 93.6 (T2I) on Flickr30K. Finally, qualitative evaluations demonstrate that FLAT representations natively support linear interpolation, latent space arithmetic, and zero-shot composed retrieval.
Read the full abstract
Computation and Language
Challenges of Auditing: Variability in Outputs of Large Language Models for Health
People increasingly use frontier AI models for health advice, but via different access modes (e.g., ChatGPT, ChatGPT Health, APIs) with varying settings. Here, we find systematic differences across access modes. Because evaluations typically rely on APIs while consumers interact through chatbot interfaces, these discrepancies limit evaluation validity. Our findings underscore an urgent need for model providers to enable faithful replication of consumer experiences and settings for rigorous audits.
Read the full abstract
Artificial Intelligence
Do LLMs Have Values? A Quantitative Analysis and Alignment Framework for Values in Large Language Models
As Large Language Models (LLMs) increasingly handle complex subjective tasks, aligning their intentions and behaviors with human values has become a critical scientific challenge. However, current efforts are confounded by a striking behavioral paradox: they fluctuate unpredictably under minor wording changes ("swing"), yet stubbornly ignore explicit instructions to correct ingrained biases ("rigidity"). Resolving this duality is critical for reliable AI alignment. To systematically understand and safely steer these latent subjective preferences, our study is structured around three fundamental questions. First, do LLMs possess an intrinsic value system? By projecting responses from 106 LLMs (150,000 queries per model) and 95,000 human survey profiles into a shared sociological space, we empirically confirm that they do. However, they do not mirror human diversity, instead crystallizing into a highly concentrated, idealized value core. Second, how can these values be quantified? We propose the Prior-Environment-Cognition (PEC) framework. This model mathematically defines value expression as the joint outcome of inherent dispositions like parameter weights (Prior), external contexts such as user prompts (Environment), and internal reasoning processes like Chain-of-Thought (Cognition). Finally, how can LLMs' values be aligned toward a desired target? Using PEC diagnostics, we establish an adaptive "Alignment Prescription". Rather than blindly applying resource-intensive training, this method identifies the minimum effective intervention needed for each dimension, ranging from zero-cost prompts to targeted parameter updates. Extensive empirical validation confirms that our approach successfully verifies the presence of LLM values, accurately quantifies their shifts, and achieves more efficient and precise steering than conventional blind training, all without degrading general capabilities.
Read the full abstract
Signal Processing
Environment-Aware Diffusion Model for Massive MIMO-OFDM Channel Estimation
This paper proposes an environment-aware diffusion based channel estimation in massive multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems. The high dimensionality of massive MIMO channels combined with limited pilot resources makes accurate estimation challenging. To address this issue, we exploit the spatial variability of wireless channels by training a diffusion model to learn the location-conditioned distribution of channel state information, which provides an environment-aware prior for channel estimation. Based on this learned prior, a posterior inference algorithm is developed to incorporate pilot observations into the reverse diffusion process, enabling Bayesian channel estimation by combining the received-signal likelihood with the learned channel prior. By jointly leveraging location information and measurement data, the proposed approach improves estimation accuracy under limited pilot resources. Simulation results based on ray-tracing channel datasets demonstrate that the proposed method consistently outperforms conventional estimators and existing learning-based approaches across various signal-to-noise ratios and pilot configurations.
Read the full abstract
Strongly Correlated Electrons
Hartree-Fock Density-Matrix Renormalization Group for Very Long Chains
We formulate Hartree–Fock as a density-matrix renormalization group (DMRG) sweep of a single Slater determinant, assuming a density–density electron–electron interaction. We compress the occupied state and long-range interaction separately, reducing each sweep step to a small local Fock problem. Exchange is retained throughout. Uniform antiferromagnetic hydrogen chains provide an almost ideal setting for this approach, while retaining the Coulomb interaction and its $1/r$ tail. We initialize and converge a chain of 100,000 electrons in 1.8 million spatial basis functions in about two hours on a 64-GB Mac mini, using less than a third of its memory. The work per sweep is nearly proportional to chain length, and the number of sweeps needed for convergence changes little from 1000 to 100,000 atoms. DMRG's compression of states and environments thus makes large HF calculations with long-range Hartree and exchange practical on an ordinary desktop computer.
Read the full abstract
Robotics
ProxiDex: Learning Dynamics-Guided Proximity Policy for Dexterous Manipulation
Multi-finger dexterous manipulation relies on stable hand-object interactions, yet these interactions are partially observable in practice. Visual observations are often occluded by the hand, tactile sensors introduce hardware-specific modalities and calibration burdens, and existing policies rarely model how these cues evolve under actions, making them brittle under contact uncertainty. To address these, we present ProxiDex, a dynamics-guided proximity policy framework that treats hand-object proximity as an interaction state for dexterous manipulation. ProxiDex reconstructs interaction point clouds and converts geometric distances into proximity cues, forming a hardware-agnostic contact representation that provides immersive feedback during VR teleoperation. Built on this representation, ProxiDex learns action-conditioned proximity dynamics with a coupled forward-inverse design: future observation latents are predicted from actions, while proximity variations are decoded from latent changes. Leveraging these dynamics, ProxiDex adaptively reweights proximity tokens across manipulation phases and uses dynamics-consistency supervision to guide policy inference, stabilizing action generation under unreliable visual feedback. Simulation and real-world experiments demonstrate improved success rates and robustness over representative baselines across standard, unseen objects, and perturbation scenarios. Additional visualizations are available at https://proxidex.github.io/.
Read the full abstract
Nuclear Theory
The Fock representation of the Coulomb interaction. I. Two-body problem
In the momentum-space few-body theory of light nuclei, where the Faddeev equations with separable two-body input close on one-variable amplitudes, the unscreened Coulomb interaction is exactly separable by the stereographic projection of Fock, with analytic form factors and strengths, and the truncation of the expansion converges at negative energies. It is derived here as an operator identity for the potential and added, without screening or fitting, as a block of separable terms to the separable nuclear input of a charged pair. The representation is compared with the screening–renormalization method and with Coulomb-distorted form factors on two-body benchmarks: the off-shell $t$~matrix, the Coulomb displacement of the $\alpha p$ Pauli-forbidden state, and the Coulomb-modified phases of the $pp$, $\alpha p$ and $\alpha\alpha$ pairs. The Feshbach–Schur projection of the Pauli-forbidden state is shown to require the Coulomb-dressed eigenstate, which the Fock block provides within the same separable problem. At positive energy a finite truncation is a screened Coulomb potential with an analytic, momentum-dependent radius.
Read the full abstract
High Energy Astrophysical Phenomena
Supernova nucleosynthesis: a review
Supernovae are major drivers of cosmic chemical evolution. They synthesize heavy elements and disperse them into the interstellar medium via their explosion shocks. Light elements are converted into heavier ones during both presupernova evolution and the explosive event itself. Supernova explosions generate nucleosynthesis environments rich in neutrons, protons, and neutrinos under unique thermodynamic conditions, which can enable the production of heavy elements beyond iron. Modern numerical simulations are constructing increasingly realistic explosion models of various supernova channels, providing more accurate nucleosynthesis conditions and chemical yields. In tandem with advances in large-scale spectroscopic surveys delivering precise, high-resolution stellar abundance data, as well as isotopic ratios from presolar grains and meteorites, our understanding of the supernova role in cosmic nucleosynthesis is poised to advance significantly. We review recent progress in modeling various supernova channels, with particular emphasis on nucleosynthesis yields derived from state-of-the-art simulations. We examine the roles of Type Ia, core-collapse, electron-capture, and pair-instability supernovae in producing intermediate-mass, iron-peak, trans-iron, and very heavy elements, as well as their characteristic chemical imprints. We also outline major theoretical uncertainties that affect yield predictions. We intend this review to serve as a timely reference for theoretical model development and a practical guide for interpreting observational abundance data.
Read the full abstract
Soft Condensed Matter
Adsorbing Boundaries Stratify the Kinetics of Confined Colloidal Phase Separation
Phase separation in network-forming materials couples mass transport to mechanical relaxation, yet how adsorbing boundaries reshape this coupling remains unclear. We compare fluid-particle-dynamics simulations resolving solvent hydrodynamics with free-draining Brownian dynamics for attractive colloids confined between parallel walls. Adsorbing boundaries do not slow coarsening uniformly but stratify the network into spatially distinct kinetic regimes. With hydrodynamics, the self-similar central network exhibits growth consistent with poroelastic relaxation, Lb~t^(1/2), whereas the adsorbed region follows an effective Lw~t^(1/3) law. In free-draining dynamics, central coarsening approaches t^(1/3), while the boundary exponent remains time dependent over the accessible times. Direction-resolved displacements reveal suppressed wall-normal motion and reduced lateral mobility near the walls. Increasing wall attraction connects adsorbed islands into an extended layer while leaving local dense-phase packing nearly unchanged. For island-like domains, we propose a testable encounter-controlled scaling framework linking boundary growth to domain geometry, size-dependent mobility, and evolving adsorbed mass. Locally self-similar central coarsening thus coexists with distinct or nonstationary boundary kinetics, precluding a common dynamic-scaling length for the central and adsorbed regions over the accessible times; in both regions, the growth law depends on whether solvent hydrodynamics is resolved.
Read the full abstract
Sound
CLASH: Counterfactual Auditing of Lexical and Prosodic Reliance in Spoken Sarcasm Detection
Spoken sarcasm detectors may exploit lexical content, prosody, or their interaction, yet conventional evaluation cannot reveal which cues drive their predictions. We introduce CLASH (Controlled Lexical-Acoustic Separation Harness), a bilingual counterfactual diagnostic framework that evaluates each utterance under original, lexical-preserving, prosody-preserving, and approximately neutralised conditions. We evaluate handcrafted acoustic-feature systems, self-supervised learning (SSL) probes, and large audio language models (LALMs) on CMMA and MUStARD. For target-only Qwen3-Omni, lexical-preserving speech retains a 0.135–0.148 AUROC advantage over prosody-preserving speech after duration balancing, with cluster-bootstrap intervals above zero; alternative lexical resynthesis preserves this advantage. Acoustic interventions shift scores without consistently improving discrimination or changing binary predictions under the evaluated conditions. Context and interaction estimates vary across corpora. These findings distinguish acoustic sensitivity from sarcasm discrimination while exposing duration, identity, and transformation effects.
Read the full abstract
Human-Computer Interaction
EmoPhone: A Multi-Wave Dataset for In-the-Wild Mobile and Wearable Affect Sensing
We introduce a three-wave, in-the-wild multimodal dataset for affect sensing that integrates smartphone sensing, wearable sensing, and dense experience-sampling-method (ESM) labels collected annually from 2020 to 2022. The dataset supports moment-level affect modeling through a shared dimensional label core across all waves, with additional affective descriptors available in the third wave (D-3). We describe the resource in terms of study design, temporal density of in-situ labels, and sensing and label coverage across waves. To support evaluation within this resource, we define an initial three-setting benchmark spanning temporal prediction from within-user history, within-wave cross-user generalization, and cross-wave generalization in which each wave is treated as a separate dataset. Our benchmark results show that the strongest method family depends on the evaluation setting: supervised baselines perform best in the temporal setting, unsupervised domain adaptation is strongest overall in the within-wave cross-user setting, and domain generalization shows the strongest overall cross-wave performance, although its margin over strong baselines is modest. These findings indicate that robust mobile affective computing is constrained not only by label availability but also by substantial participant-level variability and realistic cross-wave differences inherent in longitudinal in-situ deployments.
Read the full abstract
Algebraic Geometry
Nilpotent Jacobian maps in dimension three and stable tameness in block extensions
We study polynomial maps in three variables with nilpotent Jacobian over a field of characteristic zero. The proposed classification reduces maps with linearly independent components to a family determined by a univariate polynomial evaluated at a quadratic coordinate. The geometric part of the argument produces two algebraically dependent constant linear combinations of the components. A derivation argument then yields the normal form over the original field. We obtain explicit polynomial inverses and tame factorizations. We then study higher-dimensional maps in which all but the last three components depend on three variables. For this class, we give separate normal forms for the two three-variable blocks and deduce stable tameness from the residue-field criterion of Berson, van den Essen, and Wright.
Read the full abstract
Machine Learning
Recovering Governing Dynamics from Distributed Observations via Exact Spline Merging
Scientific observations are frequently distributed across locations, time periods, and institutions. Combining such observations into a continuous, differentiable field enables recovering governing physical parameters from its derivatives. This paper makes two contributions in this setting. First, the established additive structure of fixed-basis ridge-regression statistics is applied to tensor-product spline fields: each data holder computes a local Gram matrix and moment vector, and the merged solution is mathematically identical to centralized fitting, with no raw data shared and no iterative synchronization. This property is specific to the fixed-feature squared-error setting; the present derivation does not establish an analogous guarantee for general jointly trained multilayer networks. Second, a complete pipeline connects distributed observations to physical parameter inference through field reconstruction, derivative extraction, and linear regression. The pipeline is validated on four PDEs: diffusion, wave, heat-with-source, and the nonlinear viscous Burgers equation, recovering governing parameters to sub-percent accuracy in the linear cases and 5% for Burgers. In all cases, distributed merging introduces zero degradation relative to centralized fitting. Synthetic experiments validate parameter recovery; application to 41 years of NOAA sea-surface temperature data validates field reconstruction and aggregation equivalence on real spatiotemporal observations. Source code to reproduce all experiments is available at https://github.com/NAVEENMN/splinemerge.
Read the full abstract
Computer Vision
GraLoD: Graphics-Inspired Continuous Level-of-Detail Learning for Image Restoration
The spatial support required for image restoration varies across degradation types, image regions, and reconstruction stages. However, most existing methods rely on predefined multi-scale hierarchies and aggregate features through fixed fusion or attention, leaving the representation scale itself largely determined by the network architecture. This limitation becomes more pronounced when a task-specific backbone is extended to heterogeneous degradations in all-in-one restoration. Inspired by level-of-detail (LOD) rendering in computer graphics, we propose GraLoD, a plug-and-play framework that treats restoration scale as a spatially varying and stage-dependent continuous variable. GraLoD reuses the native encoder hierarchy, aligns its multi-scale features into a shared LOD representation space, and predicts a stage-conditioned LOD field at each decoder stage. Each spatial location then continuously queries only two neighboring representation levels, enabling the effective restoration scale to adapt to both local image content and reconstruction progress. To prevent degenerate or arbitrary scale selection, we further introduce minimal-sufficient footprint calibration (MSFC) together with structure-aware regularization (SAR) to encourage restoration-effective and spatially coherent LOD assignments. GraLoD can be directly integrated into existing restoration backbones without redesigning their fundamental feature-processing blocks. Extensive experiments demonstrate consistent improvements in task-specific and all-in-one restoration.
Read the full abstract
Materials Science
Nonlinear Photocurrent Spectroscopy and Polarization-Tunable Shift Current in the Layered Semiconductor $\mathrm{CuScP_2S_6}$
The class of layered metal thiophosphates and selenophosphates are emerging as promising candidates for non-linear photonic and optoelectronic applications. Here we report one-photon absorption (1PA) and two-photon absorption (2PA) induced photocurrents in $CuScP_2S_6$, establishing it as an optoelectronically-active non-centrosymmetric semiconductor. Spectrally-resolved 1PA and 2PA photoresponsivity paired with photoluminescence and differential reflectance measurements reveal a band gap of 2.35 eV, several higher energy interband transitions, and sub-band gap multiphoton excitonic transitions. Vertical graphite/$CuScP_2S_6$/graphite devices exhibit zero-bias photocurrent with a polarity controlled by both photon energy and optical polarization orientation within the ab-plane. This sign-dependence combined with a characteristic power-law transition from linear to square-root scaling are consistent with a shift-current origin to this bulk photovoltaic effect (BPVE). These findings establish $CuScP_2S_6$ as a multi-functional platform for polarization-sensitive non-linear optoelectronics.
Read the full abstract
Dynamical Systems
Non-uniqueness of characteristic curves of folding rational maps
A folding rational map is a postcritically finite rational map $f$ admitting an essential Jordan curve $\beta$, disjoint from the postcritical set $P_f$, whose preimage is disconnected and consists of Jordan curves homotopic to $\beta$ relative to $P_f$. Such $\beta$ is called a characteristic curve of $f$. We construct a real postcritically finite rational map $f$ of degree $6$ with $\#P_f=4$, which admits two non-homotopic characteristic curves $\beta$ and $\beta'$. It is hyperbolic with orbifold signature $(2,2,3,\infty)$. This non-uniqueness is not a Lattès phenomenon, nor does it arise from a mating decomposition. The map is constructed piecewise, and a numerical model is provided to illustrate its dynamics. Thurston obstructions are excluded by arc-lifting and intersection-number arguments.
Read the full abstract
Plasma Physics
KEMPIC-3D: A transparent and extensible electromagnetic Particle-in-Cell framework for kinetic plasma simulations
KEMPIC-3D, a fully electromagnetic three-dimensional particle-in-cell (PIC) code, has been developed in C/C++ and systematically verified for the self-consistent simulation of kinetic plasma phenomena. The numerical framework combines a staggered Yee finite-difference time-domain solver for Maxwell's equations with a relativistic Boris particle pusher, trilinear field interpolation, and charge-conserving current deposition. Its modular architecture emphasizes algorithmic transparency and extensibility, allowing the principal numerical routines to be directly inspected and modified to incorporate additional physical models and computational capabilities. The implementation is assessed through a hierarchy of benchmark problems addressing both the individual numerical components and the self-consistent particle–field coupling. The particle pusher is verified against analytical solutions for relativistic charged-particle motion, while the electromagnetic field solver is evaluated through guided-wave propagation and grid-convergence analysis. The complete PIC algorithm is subsequently verified through plasma oscillations, global energy conservation, and discrete charge continuity, demonstrating the numerical accuracy, conservation properties, and self-consistency of the coupled formulation. The capabilities of KEMPIC-3D are further demonstrated through a representative simulation of laser-driven plasma wakefield generation in an underdense plasma. The simulations reproduce the characteristic nonlinear wakefield structure and exhibit stable numerical behavior across different particle discretizations without additional spatial filtering. Together, these results establish KEMPIC-3D as a reliable, transparent, and extensible computational framework for multidimensional kinetic plasma simulations.
Read the full abstract
Digital Libraries
Testing Our Foundations: Citation Trends, Errors, and Emerging Hallucinations in the Computing Education Literature
Accurate references are foundational to scholarly work, enabling verification, attribution, and systematic review. However, the rapid adoption of large language models has introduced a serious integrity concern: plausible-looking but fabricated citations. Although hallucinated references are widely discussed, their visibility within specific research communities remains unclear. We address this gap by examining reference integrity at key computing education venues using ACM Digital Library data. We analyze referencing trends across 24,751 computing education papers and compare them with the broader ACM corpus of more than 723,000 papers and 15 million references. We then examine reference lists from these venues, classify common bibliographic errors, and manually identify LLM-generated hallucinations containing verifiably false information, including impossible page ranges, invented titles, and misattributed authors. In 2025, hallucinated references appeared across five SIGCSE-sponsored or in-cooperation venues. At the Technical Symposium alone, verified hallucinated references increased from 3 in 2025 to 17 in 2026, appearing in 2.3% of 2026 proceedings papers. Although still relatively rare for now, this growth poses an integrity risk our community should not ignore.
Read the full abstract
Machine Learning
AsyncCouple-Flow: Asynchronous Cross-Modal Coupling and Flow Matching for Spatio-Temporal Forecasting
Multi-modal spatio-temporal forecasting (MM-STF) supports weather nowcasting, traffic prediction, and earth-system modeling by combining heterogeneous sources such as physical fields, satellite imagery, and in-situ sensors. Three obstacles persist: (i) modalities have different spatio-temporal sampling rates, forcing lossy interpolation onto a unified grid; (ii) modalities are frequently missing at deployment due to sensor outages or revisit gaps, while most methods train with full availability; and (iii) autoregressive decoders accumulate errors over long horizons, amplified by multi-modal conditioning. We propose AsyncCouple-Flow to address these issues jointly. A Modality-Aware Token Sparsification (MATS) module performs scale-aware tokenization and uses a shared importance scorer to select top-k tokens per timestep, producing equal-length sequences. An Asynchronous Cross-Modal Coupling Graph (ACCG) replaces fixed cross-attention with a learnable graph whose edges encode time offsets, semantic similarity, and modality-specific physical priors, enabling fusion under arbitrary asynchrony and missingness. A Flow-Matching Forecasting Head models multi-step prediction as a conditional ODE, trained with stochastic modality dropout and integrated jointly to avoid autoregressive drift. Experiments on ERA5+GOES+ISD weather forecasting and PEMS-BAY traffic prediction with multi-source side information show that AsyncCouple-Flow outperforms state-of-the-art baselines and remains robust with up to two missing modalities. The code will be released upon acceptance.
Read the full abstract
Computer Vision
Efficient Text-to-Image Generation: An Adaptive Step Schedule Controller for Diffusion Models
Text-to-image diffusion models often use a fixed number of denoising steps, balancing time costs and image quality. However, the optimal number of steps depends on the complexity of the input text prompt. We propose an adaptive diffusion controller that dynamically adjusts the number of steps to generate high-quality images efficiently, without additional model training. By leveraging a mixture of step schedules with varying step sizes and evaluating the error term discrepancy at each timestep, our method transitions between schedules to optimize performance. Experiments on COCO and DiffusionDB show that our approach reduces inference time while maintaining visual fidelity, offering a more efficient alternative for text-to-image diffusion models.
Read the full abstract
General Relativity and Quantum Cosmology
Black-Hole First Laws and Horizon Constraints: A Differential Rank Criterion
Thermodynamic rewritings of horizon equations depend on which relations are varied and which have already been imposed. We formulate a differential rank criterion that makes these choices explicit. The construction uses a stationary metric family together with an independently marked radial surface, physical charges determined by the metric parameters, and specified extensions of the horizon thermodynamic quantities. A first-law covector is pointwise representable by selected constraints precisely when it lies in the span of their differentials. We distinguish this algebraic condition from a smooth identity in a neighborhood. If regular constraints already define the complete physical family, pointwise representability follows from the first law itself; a neighborhood identity additionally depends on the chosen extensions. With explicit prescriptions, Kerr–Newman, rotating BTZ black holes in new massive gravity, and the new-type NMG black hole all exhibit exact alignment with their horizon constraints. By contrast, substituting an exact metric family into a field-equation component can annihilate that residual before any horizon condition is tested. We also give an explicit entropy–volume comparison and specify the role of fixed ensemble data in free-energy variations. The framework separates consequences of the established first law from additional claims about selected constraints and their extensions.
Read the full abstract
Materials Science
Origin and Reduction of Coercive Fields in ZnO-based Wurtzite Ferroelectrics
Wurtzite ferroelectrics offer a route to semiconductor-compatible non-volatile devices, but their large coercive fields (E$_c$) constrain the operating window between polarization reversal, leakage and dielectric breakdown. Atomically localized and non-classical switching fronts have recently been identified in wurtzite nitrides, yet the microscopic feature that sets the coercive field remains unresolved. Here we use large-scale reactive molecular dynamics, parameterized against first-principles data, to follow field-driven reversal in pristine, Mg-modified and heterostructured ZnO. Reversal proceeds through rugged inversion-boundary filaments in agreement with recent observations in other wurtzite ferroelectrics. We find that the field required for switching is controlled by the advancing filament head, which is a under-coordinated reconstructed inversion-boundary region during the transient switching dynamics with a large field-aligned local polar-order amplitude that is only partially compensated by an oppositely oriented surrounding shell. Controlled strain-only and charge-doping-only perturbations show that charge redistribution lowers the coercive field more effectively than strain because it suppresses this field-setting unscreened local polar order at the filament head, with the combined effect lowering E$_c$ by $\sim$ 50 %. Layered ZnO/ZnMgO architectures further create buried nucleation sites and shorten filament propagation lengths. These results connect atomistic switching topology to a materials-design principle for substantially reducing coercive fields in ZnO-based wurtzite ferroelectrics.
Read the full abstract
Systems and Control
Joint Freshness and Age-Dispersion Control over Finite-State Markov Wireless Channels
Age of information (AoI) has become a standard design objective for timely monitoring as it measures the freshness of the latest update at a receiver. AoI alone, however, is insufficient in goal-oriented applications where decisions depend on consecutive observations. Age dispersion complements AoI by measuring the generation-time separation between consecutive updates. In this paper, we study joint freshness and dispersion control for a generate-at-will status update link modeled as a finite-state Markov wireless channel. The objective is to minimize the long-run probability that either AoI or age dispersion exceeds a prescribed threshold under an average transmission rate constraint. For channel-dependent randomized transmission policies, we derive exact matrix expressions for the stationary joint distribution of AoI and dispersion. For reversible channels, mean dispersion is lower bounded by the reciprocal of the delivery throughput, and the difference is an explicit non-negative variance term for which we establish the exact equality condition. We then formulate the adaptive joint-threshold control problem as a constrained Markov decision process and prove that it admits an exact finite-state representation whose size scales linearly with the number of channel states and the threshold levels. Simulation results show that the controller reduces joint threshold violations by 37.3% relative to adaptive AoI-only control under the same transmission budget, while increasing mean AoI by 8.3%.
Read the full abstract
General Relativity and Quantum Cosmology
Thermodynamic topological classification of magnetically charged slowly rotating Kerr black holes in nonlinear electrodynamics with a cosmological constant
We provide an explicit rotating black-hole realization of the predicted \(\overline{W}^{1-}\) thermodynamic topological subclass by studying a slowly rotating, magnetically charged Kerr black hole in nonlinear electrodynamics (NLED). Using a reduced fixed-parameter off-shell potential, we derive the inverse-temperature curve analytically rather than relying solely on numerical plots. In AdS spacetime, this curve begins at a finite nonzero value, increases monotonically, and diverges at large horizon radius; it therefore contains exactly one defect with \(w=W=-1\) and has no stable large-black-hole branch. Positivity of the analytic coefficient establishes this result throughout the admissible slow-rotation parameter domain. Under the same reduced prescription, the Kerr-AdS benchmark instead has two branches and belongs to the \(W^{0-}\) class, establishing a clear topological distinction between the NLED-Kerr and classical rotating cases. In dS spacetime, the formally extended off-shell curve remains monotonic and any isolated zero has \(w=-1\); because the parameter-dependent horizon-admissible range is not fixed in the present reduced analysis, no physical dS asymptotic subclass is assigned.
Read the full abstract
Computer Vision
Counterfactual Reasoning for Robust Visual Question Answering
Modern Visual Question Answering (VQA) models often exploit spurious correlations in training data, leading to poor out-of-distribution (OOD) generalization due to language bias. Although counterfactual learning has shown promise, existing methods can be improved to better guide attention toward causal evidence and strengthen feature discrimination. To address this, we propose a novel training framework that enhances counterfactual contrastive learning for VQA. Our framework introduces three key contributions: (1) a three-stage curriculum for stable multi-objective optimization, (2) an enhanced Batch-Contrastive loss for more discriminative feature learning, and (3) two novel regularizers, Answer-Contrastive (AC) loss to refine the prediction space and Gradient-Discrepancy (GD) loss to enforce causal visual grounding. Our model achieves a competitive accuracy of 61.64% on the bias-sensitive VQA-CP v2 benchmark while maintaining 62.80% on the standard VQA v2 dataset, yielding a small generalization gap of 1.16%. This demonstrates a strong balance between OOD robustness and in-distribution performance.
Read the full abstract
Probability
On Completion Times under Memoryless Catastrophe
We study the completion time of a task subject to independent reset (catastrophe) at each step. The completion-time PGF depends on the base-process PGF through an affine relation, and we exploit this structure systematically. Our main result shows that, among age-based catastrophe mechanisms, geometric-tail catastrophe is exactly the class that yields uniform affine PGF structure; in continuous time, the characterization sharpens to Poisson resetting. We establish a sharp two-sided Kolmogorov bound of order $p+|\alpha|$ for the exponential approximation $d_K(T/E[T], \mathrm{Exp}(1))$, thereby closing a logarithmic gap. Applications to the coupon collector with reset coupons reveal a discontinuous Gumbel-to-Exponential transition under resetting, while a multi-phase model exhibits a Gaussian-to-exponential transition with exponential convergence rate.