Information Retrieval
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Monday 14 September
Our team, VANGUARD, presents IROH (Insightful Ranking of Humor), a three-stage retrieval system for JOKER Task 1 English at CLEF 2026, achieving first place on the leaderboard with 0.6347 MAP. Our pipeline combines hybrid sparse-dense retrieval, cross-encoder reranking, and a LoRA-adapted Large Language Model judge ensemble. We employ Gemma 4 to generate query-aware rationales under two prompt strategies, generic and typed, and produce up to four types of structured hard negatives for training data construction. Through an ablation across three cross-encoder architectures, four dense embedders, and eight judge configurations, our key findings are threefold: (1) the rationale-distilled judge is the primary driver of ranking quality, whereas appending rationales to the first-stage index contributes negligibly; (2) structured hard negatives degrade generalisation in nearly all configurations despite inflating local validation scores; and (3) across the components we ablate, the lighter, better-calibrated model is competitive with or stronger than its larger counterpart, with the generic-rationale Qwen2.5-7B judge (0.6055 MAP) outperforming every Gemma-4-31B configuration, and the advantage of generic over typed rationales is concentrated almost entirely in the smaller model.
Dynamical Systems
Jet rigidity of Mather's $\beta$-function and complexified KAM curves holomorphic in potential for analytic standard maps
Mather's $\beta$-function associates with each rotation number the least average action carried by an orbit of a twist map. For analytic standard maps in the KAM regime, we prove the following. For any family of $d \ge 1$ directions of perturbation satisfying a natural symmetry condition, the jet of the $\beta$-function at a single algebraic Diophantine rotation number locally determines the potential for an open and prevalent, hence dense, set of base potentials in the KAM domain. In particular, we obtain that for a residual and prevalent set of even potentials in the KAM domain, every real analytic deformation with finite Fourier support that preserves the full jet of the $\beta$-function at a fixed algebraic Diophantine rotation number is constant. These results are non perturbative within the KAM domain. As an intermediate result of independent interest, we establish a KAM theorem giving joint $C^\infty$–holomorphic dependence of the invariant curves, and hence of the $\beta$-function, on complex holomorphic potentials and on a domain of complexified rotation numbers whose boundary contains real Diophantine numbers.
Read the full abstract
Audio and Speech Processing
OLAC: An Overlapped Lossless Audio Codec in the Time-Domain with MDCT Compatibility
Lossless audio coding is a highly mature field of research, with limited potential for significant improvements in pure compression performance. However, emerging real-time wireless applications increasingly require dynamic transitions between lossy and lossless coding to adapt to fluctuating network capacities. Existing standalone lossless codecs cannot achieve this seamless switching without discontinuity. In this paper, we propose a lossless codec based on time-domain aliasing cancellation (TDAC) that can match the overlap in the CELT mode of the Opus codec. This allows the transition between lossy and lossless coding to be achieved without discontinuity or the transmission of redundant information. We show that the proposed codec still achieves state-of-the-art lossless compression without being penalized by its use of TDAC.
Read the full abstract
Instrumentation and Methods for Astrophysics
Calibrations for high precision differential astrometry onboard Theia and HWO
Future space astrometry missions such as Theia and an astrometric mode of the Habitable Worlds Observatory (HWO) require detector calibration and instrumental characterization at an unprecedented level to achieve sub-micro-arcsecond precision. This work presents three complementary developments addressing these requirements. First, we report an independent characterization of the GIGAPYX-4600 back-side illuminated CMOS detector, evaluating its linearity, readout noise, dark current, pixel response non-uniformity, defective pixel fraction, and inter-pixel capacitance. The results demonstrate excellent detector performance and confirm its suitability as a candidate for future gigapixel focal planes. Second, we develop an interferometric calibration method based on Young's fringes to measure pixel centroid displacements, enabling the characterization of inter-pixel response variations with an approach compatible with onboard implementation. The method is investigated through numerical simulations and an experimental testbed using the GIGAPYX-4600 detector. Finally, we introduce an astro-calibration framework that jointly estimates stellar astrometric parameters, telescope attitude, plate scale, and optical distortion through a global iterative optimization. Although this calibration approach is still under development, simulations and laboratory activities are underway to validate its performance. Together, these developments contribute to the technological and calibration framework required for the next generation of high-precision astrometric space missions.
Read the full abstract
Cryptography and Security
Tractable Defense against Advanced Persistent Threats in Networked Settings
Recently, the theory of Boolean Dynamical Systems was proposed to study the decision theory surrounding the defense of computer networks against Advanced Persistent Threats (APTs). Boolean Dynamical Systems naturally capture four first principle primitives of APTs: the stealthy nature of attacks, limited and noisy information from automated systems like intrusion detection systems, lateral movement after the attacker penetrates into the network, and the defender's ability to secure a subset of computers at any time at the loss of resources such as system uptime. Currently, doing optimal/heuristic control in a computationally tractable manner is not possible because the emergent value function is computationally intractable (with respect to the network size). To resolve this, we propose a mean-field analysis inspired heuristic value function. We prove that our proposed heuristic is based on an exact computation of the value function under the assumption that the underlying state estimate distribution maximizes entropy. We numerically evaluate the quality of our heuristic as parameterized by the degree to which the entropy assumptions are violated.
Read the full abstract
Algebraic Geometry
Small-Subgroup Criteria for Liftability of Automorphism Groups of Smooth Hypersurfaces
In this paper, building on our previous Sylow criteria, we establish small-subgroup criteria of liftability and $F$-liftability for the linear automorphism group $G$ of smooth hypersurfaces $X$ over algebraically closed field of characteristic zero. When $\mathrm{dim} X = p-2$ for some odd prime $p$, we prove that ($F$-)liftability of any finite subgroup of $G$ can be tested on its $p$-subgroups of order at most $p^2$. When $X$ is a degree $p$ hypersurface of dimension $2p-2$, we prove that ($F$-)liftability of $G$ can be tested on all its $p$-subgroups of order at most $p^3$, which is sharp for $p\geq5$. When $p=3$, this bound improves to $9$, giving the corresponding criteria for smooth cubic fourfolds.
Read the full abstract
Quantum Physics
Reducing Decoding Latency in Quantum Error Correction by Early Starting Clustering
In quantum error correction, fast low-latency decoding is essential for fault-tolerant quantum computation, as delays in processing syndrome data can lead to the backlog problem. Existing decoders, including parallelizable approaches such as Union-Find, begin decoding only after all stabilizer measurement outcomes from an error-correction cycle have been received, inherently introducing a delay before decoding begins. We introduce Cluster-As-You-Go (CAYG), a modification of the Union-Find decoder that processes syndrome information during stabilizer measurements by clustering and correcting errors as they appear. This approach reduces the size of the remaining decoding problem at the end of the quantum error correction cycle and, consequently, the time required to complete the decoding process. While this early start of clustering incurs a modest reduction in decoding accuracy, it preserves the decoder's scalability. Surface-code simulations show that the resulting reduction in post-measurement idling can outweigh the accuracy loss, yielding an improved speed-accuracy trade-off. These results demonstrate that real-time, early-starting decoding during QEC cycles is both feasible and can be advantageous for quantum error correction. Demonstrated here for the surface code, CAYG is broadly applicable to other quantum error-correcting codes, which allow for clustering-based decoding approaches, and is extensible to dedicated real-time decoding hardware.
Read the full abstract
Machine Learning
Multi-View Molecular Representation Learning with Hierarchical Graphs and Contextualized Fingerprints
Molecular property prediction requires representations that generalize from limited labeled data to structurally novel compounds. Existing molecular pretraining methods often rely on a single view: graph-based approaches model atom-bond topology but provide limited fragment-level supervision, whereas fingerprint descriptors encode chemical patterns but are typically used as fixed auxiliary features. We propose HiFi-Mol, a multi-view framework that separately pretrains a hierarchical graph encoder and a contextualized fingerprint encoder before downstream integration. The graph branch uses fragment-aware masking with multi-resolution supervision to capture substructure-aware representations, while the fingerprint branch tokenizes active entries from seven fingerprint families and applies masked language modeling to learn contextualized embeddings. During fine-tuning, HiFi-Mol combines projected multi-resolution graph features with fingerprint embeddings for downstream prediction. Evaluated on MoleculeNet benchmarks under the scaffold split, HiFi-Mol achieves a 2.77% improvement in average ROC-AUC over the best baseline across eight classification tasks while maintaining competitive performance on three regression tasks. Further analyses reveal that fragment-aware masking improves graph representation quality, and classification results demonstrate dataset-dependent strengths of the individual graph and fingerprint variants, confirming that the two views provide complementary predictive signals.
Read the full abstract
Classical Analysis and ODEs
BMO-Type characterizations and endpoint estimates for commutators of intrinsic square functions on Orlicz-Hardy spaces
The commutators of intrinsic square functions associated with the intrinsic Littlewood–Paley $g$-function, the intrinsic $g_{\lambda}^{*}$-function and the intrinsic Lusin area function can be used to characterize BMO-type function spaces through their boundedness properties. In this paper, we show that, for $b\in {\rm BMO}(\mathbb{R}^{n})$, the boundedness of each of these commutators from the Musielak–Orlicz Hardy space $H^{\varphi}(\mathbb{R}^{n})$ to $L^{\varphi}(\mathbb{R}^{n})$ is equivalent to $b\in {\rm BMO}_{\varphi}(\mathbb{R}^{n})$ (a nontrivial subspace of $\rm{BMO}(\mathbb{R}^{n})$), under suitable assumptions on the growth function $\varphi$. This extends the weighted characterization of Han and Wu [Proc. Amer. Math. Soc. 152(1) (2024), 281–293] to the Musielak–Orlicz setting. In addition, under suitable assumptions on a growth function $\Phi$, we prove that, for $b\in {\rm BMO}_{\Phi}(\mathbb{R}^{n})$, these commutators are bounded from the $b$-adapted Orlicz–Hardy space $H_{b}^{\Phi}(\mathbb{R}^{n})$ to $L^{1}(\mathbb{R}^{n})$ and from Orlicz Hardy space $H^{\Phi}(\mathbb{R}^{n})$ to $L^{1,\infty}(\mathbb{R}^{n})$.
Read the full abstract
Artificial Intelligence
Diversified and Perceptible Counterfactual Examples Leveraging Expert Knowledge
CounterFactual Examples (CFEs) are a cornerstone of eXplainable Artificial Intelligence (XAI), offering local, post hoc, and model-agnostic explanations by identifying minimal input modifications that alter a model's prediction. Yet, in order to be intelligible, these modifications must also be semantically meaningful to the explainee. This paper proposes to integrate knowledge expressed as a fuzzy linguistic vocabulary to represent the explainee's perception and interpretation of the data. The domain induced by this fuzzy vocabulary imposes structural constraints that make the features dependent, preventing the use of gradient-based optimisation methods for CFE generation, e.g., DiCE. The paper proposes a continuous data embedding in this linguistic domain and exploits it to define a variant of DiCE that allows personalisation for the explainee, named DiCEf. As illustrated by experimental results on a real-world dataset, this extension of the DiCE method enables the generation of CFEs that are linguistically perceptible while preserving cost minimality, sparsity, and diversity.
Read the full abstract
Computation and Language
Through the Eyes of the Beholder: Biometric and Demographic Conditioning for Multimodal Sexism Detection
Detecting sexism on the internet is a fundamentally subjective task; our team, VANGUARD, addresses this challenge in the EXIST 2026 Task 2 by proposing a human-centered multimodal framework that analyses and incorporates the psychological and demographic characteristics of human annotators into the detection pipeline. We fuse five input modalities through a cross-attention architecture with Feature-wise Linear Modulation conditioning. Meme text is extracted and visually described with Gemma 4, then augmented by automatic translation between English and Spanish with NLLB-200. Text and image representations are produced by LoRAadapted XLM-RoBERTa and CLIP encoders and fused with sensor features encoded by a pretrained autoencoder. To model annotator subjectivity, we frame Subtask 2.1 as a label distribution learning problem, optimizing a Kullback-Leibler divergence loss over the full annotator label distribution. At inference time, predictions are produced by soft-voting between the deep multimodal network and a complementary SVM trained on stylometric and physiological features. Our best submission ranks 29th out of 114 on Subtask 2.2 (source intention) under soft evaluation, and the normalized ICM scores remain above the baseline on Subtasks 2.1 and 2.2, indicating that annotator-centered conditioning contributes a usable signal. We release our full pipeline and analysis to support reproducible human-centered modeling.
Read the full abstract
Analysis of PDEs
Global Sobolev Convergence of Howard Iteration for Finite-Horizon Controlled Diffusions
We study Howard policy iteration for finite-horizon stochastic control problems in which the control enters the diffusion coefficient. Each iteration selects a policy by maximizing the Hamiltonian and evaluates it through a linear parabolic equation. Since the improvement depends on the Hessian, even smooth data can produce discontinuous evaluation coefficients. We prove global convergence in the full parabolic Sobolev space under assumptions that allow this loss of regularity. In one space dimension, bounded measurable coefficients and uniform ellipticity suffice, without a large discount, a short horizon, small diffusion variation, or regularity assumptions on the improving policies. Multidimensional extensions hold under structural conditions on the diffusion matrix. The proof combines monotonicity with a parabolic growth estimate to show that the Bellman residual vanishes in measure. Uniform higher integrability then yields strong convergence of the values, gradients, Hessians, and time derivatives, and constructs the unique strong solution. We also establish residual error bounds, extend the argument to entropy regularization and zero-temperature limits, and obtain quadratic convergence for special one-dimensional models.
Read the full abstract
Computer Vision
VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding
Multimodal Large Language Models (MLLMs) have achieved remarkable progress on short video understanding yet remain limited on long videos due to the limited visual context window. Prevailing approaches rely on uniform frame sampling or recent coarse-to-fine agentic zooming, both of which struggle to localize sparse, decisive evidence in sufficiently long videos. We formulate long video understanding as a Sequential Evidence Acquisition (SEA) problem, in which an agent reads the video turn by turn along the temporal axis, deciding at each turn how fast to watch, what evidence to retain, when to revisit uncertain segments, and when to stop and answer. Inspired by this view, we propose VideoScout, a multi-turn reasoning agent that instantiates the SEA paradigm through adaptive reasoning pacing. Specifically, by dynamically controlling the viewing pace, VideoScout enables efficient traversal of long videos within a bounded visual context window, allowing the agent to access more video content while balancing content analysis depth with reading efficiency. To train VideoScout, we construct VideoScout-66K, a set of over 66K high-quality exploration turns from 10K answer-verified trajectories, and adopt a two-stage pipeline: cold-start supervised fine-tuning teaches the agent per-turn output format, while the Decoupled Clip and Dynamic sAmpling Policy Optimization (DAPO) algorithm performs trajectory-level reinforcement learning with a composite reward that jointly considers answer accuracy, output format compliance, and the temporal alignment between the agent's viewing progress and the teacher's answer timing measured by intersection-over-union (IoU). Extensive experiments on long video understanding and reasoning benchmarks demonstrate that our 7B model achieves strong performance compared with existing trained 7B agentic models.
Read the full abstract
Numerical Analysis
Asynchronous Jacobi and randomized Gauss–Seidel methods in shared and distributed memory: A unified convergence-rate analysis
Asynchronous iterative methods are attractive for large-scale parallel computing because they reduce synchronization and communication overhead. Existing convergence-rate analyses, however, have primarily focused on shared memory implementations, whereas distributed memory systems introduce more general and potentially inconsistent communication delays. In this work, we revisit asynchronous Jacobi and randomized Gauss–Seidel (RGS) methods for symmetric positive definite linear systems from a unified perspective. We first introduce a general asynchronous model that encompasses both shared memory and distributed memory settings and expresses the two methods through a common coordinate-update framework. We then establish linear convergence in expectation under an explicit stability condition. The resulting convergence bound depends algebraically on the delay through the quantity $\sqrt{\rho\tau}+\rho\tau$, where $\tau$ is the maximum communication delay and $\rho$ reflects the communication pattern of the underlying parallel implementation. In particular, under an appropriate scaling regime with $\rho\tau=O(1)$, the guaranteed per-iteration convergence rate has the same asymptotic order as that of synchronous RGS. These results provide a unified framework for quantifying the effect of asynchronicity on asynchronous Jacobi/RGS methods in both shared and distributed memory environments.
Read the full abstract
Quantum Physics
Parametric two-qubit gates via Landau-Zener interference
We propose and demonstrate gates between two superconducting qubits based on quantum interference of consecutive Landau-Zener (LZ) transitions. This gate mechanism bridges between baseband and parametric two-qubit control, enabling in situ tuning of the control frequency across a continuous interval up to hundreds of MHz. Another advantage compared to dispersive couplers is that the speed of the LZ gate is on the order of the full coupling strength. We experimentally demonstrate the gate on two platforms, a modular chiplet architecture of coupled generalized flux qubits, and on a monolithic transmon architecture. The combination of tunability and gate speed establishes the LZ gate as a unique tool for multiplexing control pulses and interconnecting superconducting chiplet architectures.
Read the full abstract
Computer Vision
A Unified Vision-Language Model for PSMA PET/CT Report Generation, Visual Question Answering, and Lesion Segmentation
Accurate PSMA PET/CT interpretation is central to prostate cancer management, yet existing PET/CT AI models typically address isolated tasks. We propose a unified PSMA PET/CT vision-language model for report generation, visual question answering, and lesion segmentation. The framework adopts an LLaVA-style architecture, comprising a PET/CT vision encoder, an MLP-Mixer projection module, a LoRA-tuned large language model, and a 3D segmentation branch. Training followed a four-stage strategy: vision encoder pretraining, projection-layer alignment, VLM fine-tuning, and final multitask tuning. Language tasks used 5,747 PSMA PET/CT datasets with paired reports, while segmentation used the PSMA subset of AutoPET. The model outperformed PET2REP and a CT-based baseline across standard report-generation metrics, improved performance across VQA question types, and achieved higher Dice and lesion-level overlap F1 than SegAnyPET and nnUNet. These results support the feasibility of a unified framework for structured, interactive, interpretable PSMA PET/CT analysis with voxel-level grounding within a single multitask model architecture.
Read the full abstract
Geometric Topology
Totally Geodesic Submanifolds of Teichmüller Space in Complex Dimension at Least Three
Let $N\subset \mathcal{T}_{g,n}$ be a connected complex totally geodesic submanifold. We prove that if $\dim_{\mathbb{C}}N\ge3$, then $N$ is the image of an entire Teichmüller space under a covering construction induced by a finite branched cover of marked surfaces. This answers a question of Arana-Herrera and Wright.
Read the full abstract
Geophysics
Three dimensional non-singular mollified elastic dislocation theory for extended width fault zones and inhomogeneous boundary element models
Classical elastic dislocation theory (CEDT) has two challenges when applied to faulting problems: 1) fictitious on-fault stresses not defined by ordinary integration and 2) the geometric unreality of infinitely thin fault zones. We show that both can be resolved by a mollified elastic dislocation theory (MEDT) built on Cortez blob (Cortez, 2001) mollified displacement discontinuity Green's functions, which represent deformation across spatially distributed fault zones of finite scale epsilon and produce singularity-free displacements and stresses everywhere. Analytical integration of the mollified source solution over arbitrary planar triangular elements is done with AI, and the resulting closed-form solutions allow for the calculation of non-singular stresses across geometrically complex fault systems. Further, we demonstrate the decoupling of the fault-zone width scale epsilon from the mesh length scale h, the elasticity analog of a result established for regularized viscous flow Stokeslets (Ferranti and Cortez, 2024). We use these mollified kernels to demonstrate numerically stable collocation boundary element models including spatially extended fault zones, material property variations and non-planar topography.
Read the full abstract
High Energy Physics (Phenomenology)
Effect of inelastic scalar dark matter in hidden $U(1)$ scenario after the LZ nuclear recoil
We study a hidden $U(1)_H$ extension of the Standard Model (SM) introducing two SM singlet complex scalar fields where one of them can be considered as a viable inelastic Dark Matter (DM) candidate due to its $U(1)_H$ charge being half of the charge of the other one which develops a vacuum expectation value. First we estimate a parameter region satisfying observed DM relic abundance through DM annihilation and further use these parameters to study inelastic DM-nucleus scattering to reproduce the recent high nuclear recoil energy event found by LZ collaboration. These interactions are mediated by a dark photon $(A^\prime)$ which is evolved after the spontaneous breaking of $U(1)_H$ symmetry and having a kinetic mixing with the SM photon. Hence we estimate bounds on the kinetic mixing as a function of $A^\prime$ mass and finally compare it with existing experimental bounds from different beam-dump and collider experiments. These parameters could be probed by lifetime and energy frontier experiments in future.
Read the full abstract
Complex Variables
On the uniform flatness of polynomial graphs
This paper grew out of an effort to understand the so-called uniform flatness, an important concept in the Bargmann–Fock interpolation theory developed by Varolin et al. We show that the graph of every polynomial in one complex variable is a uniformly flat algebraic curve in $\mathbb C^2$, with vanishing upper density with respect to the Gaussian weight. Combined with a result of Pingali and Varolin, this implies that such graphs are interpolating for the Bargmann–Fock space on $\mathbb C^2$.
Read the full abstract
Information Retrieval
Self-Evolving Memory for Generative Recommendation
Generative recommendation has emerged as a promising end-to-end paradigm for personalized recommendation. However, user preferences continuously evolve over time, making self-evolving an essential capability for generative recommender systems. Existing evolving strategies, such as continual retraining and distillation-based adaptation, directly update the shared model parameters using streaming interactions. Nevertheless, we find that directly applying such strategies to generative recommendation introduces a critical issue, termed evolution conflict. Specifically, heterogeneous preference shifts from different users are optimized within a fully shared autoregressive parameter space, causing dominant behavioral patterns to progressively dominate the model evolution process while underrepresented patterns become increasingly overlooked. To address this issue, we propose a self-evolving memory paradigm for generative recommendation, aiming to enable effective evolution across heterogeneous behavioral patterns. We further identify three key principles for effective self-evolving recommendation systems, including isolated memorization, reinforced evolution, and scalable application. Guided by these principles, we develop LION, a simple yet effective framework centered on a sparse Key-Value memory layer. Specifically, LION introduces sparse memory activation to isolate the evolution of different behavioral patterns, while a consolidation loss is designed to reinforce the learning of underrepresented preference dynamics during continual adaptation. Extensive experiments on diverse real-world datasets demonstrate the effectiveness of LION under various continual evolution settings (e.g., per-period evaluation, user/item group evaluation, and evolution convergence analysis). The codes are released at https://github.com/JazyJiang/Self-Evolving-Memory-for-Generative-Recommendation.
Read the full abstract
Number Theory
Long runs of integers with small prime factors and the divisor function of $n!$
Let $d$ be the divisor function, and let $K(n)$ be the least positive integer $K$ for which $d((n + K)!) \ge 2d(n!)$. Erdős, Graham, Ivić and Pomerance proved that, for infinitely many $n$, \begin{equation*} K(n) > (1/9)(\log n)(\log_{2} n)(\log_{4} n)/(\log_{3} n)^3. \end{equation*} We improve upon this by a factor of order $\log_{3} n$, which brings the bound to the same order as Rankin's 1938 lower bound for gaps between consecutive primes. The two problems are closely related, but a long prime-free interval does not by itself produce a large value of $K(n)$: what is needed is a weighted variant of the Erdős–Rankin construction. We follow the method of Erdős, Graham, Ivić and Pomerance, replacing a key estimate by an averaging argument that permits some integers to remain uncovered. This improvement was formulated and proved during a private interaction with Claude Fable 5.1, a publicly available generative-AI system; the argument is verified and presented here by the author.
Read the full abstract
Combinatorics
An improved bound on the treewidth of planar graphs excluding a grid minor
We show that every planar graph with no $t \times t$ grid minor has treewidth at most $4t +4$. This improves on the previously best known bound of $\frac{9}{2}t - \frac{11}{2}$, due to Gu and Tamaki (2012), and is within a factor $2$ of optimal. A key step in the proof is showing the following result, which might be of independent interest: Every $2$-connected plane graph $G$ with radius $d$ and faces of size at most $k$ has a tree-decomposition of width at most $\max\{3d+ k+5, 2d+2k+1\}$ such that the vertex set of every face of $G$ is contained in some bag.
Read the full abstract
Statistics Theory
Resolvable Fourier Degree and Minimax Estimation on theBoolean Hypercube
We study high-dimensional nonparametric regression on the Boolean hypercube \(\{0,1\}^d\), from a sample of size \(n\), under a Fourier–Sobolev budget, i.e., a bound \(B\) on the total quadratic influence of the regression function. The central object is the resolvable Fourier degree \(D^*(d,n)\), defined as the largest interaction order \(D\) such that the dimension of the degree-\(D\) Fourier–Walsh space does not exceed \(n\). We prove that the minimax risk over Fourier–Sobolev ellipsoids is of order \(B/D^*(d,n)\): estimation is possible up to the statistically visible interaction order, and the budget controls the residual mass carried by higher-order interactions. We then study monotone regression under the same budget. For monotone functions supported on a known set of \(s\) coordinates, with \(s\) of order \(\log n\), we obtain a matching minimax rate of order \(B/s\). Finally, we consider an unknown-support model in which the function is monotone and depends on at most \(s\) coordinates; its minimax risk is governed by the intrinsic dimension \(s\) together with the logarithmic cost of selecting the active support.
Read the full abstract
Signal Processing
AI-based Interference Mitigation for Power-domain Spectrum Sharing among LEO Satellites
Low-earth orbits (LEO) satellites have recently gained wide attention for their potential to provide high-data-rate services in desert, sea, and rural areas. Currently such systems operate in the Ku band on the down-link. However, inter-operator LEO satellites produce frequent In-Line-Events (ILE) creating co-channel interference between their waveforms. This situation will be exacerbated with the launch of more constellations being launched. To address this challenge, we propose a novel interference mitigation system based on power control. Specifically, satellites from one operator will adjust its transmit power to create a sufficient received-power gap for co-frequency beams, enabling successive interference cancellation (SIC) or AI-assisted algorithms on the user-side for interference mitigation, where AI requires a smaller power gap. We use the waveforms from Starlink and EutelsatOneweb to evaluate the designed system. Results show the designed system can significantly mitigate interference and increase the overall spectrum efficiency.
Read the full abstract
Differential Geometry
Upper bounds of the Laplace spectrum on compact Kähler manifolds
On a connected compact Kähler manifold without boundary, we derive a uniform upper bound for each Laplace eigenvalue over all Kähler metrics in a given Kähler class. This provides an affirmative answer to a question of Yau.
Statistics Theory
Nonparametric Regression on Fixed-Cardinality Subsets:Minimax Risk and Random-Design Effects
We study regression with subsets as covariates. The response is an unknown function of the input subset, and observations consist of noisy evaluations at uniformly sampled subsets, each containing exactly \(k\) items from a ground set of size \(d\). This problem arises in combination screening, bundle preference modeling, and other settings in which outcomes depend on collections of prescribed size. The fixed-cardinality constraint couples the membership coordinates, so standard product-domain notions of interaction and smoothness cannot be imported unchanged. We use the harmonic decomposition of the Johnson graph to define an intrinsic Johnson–Sobolev scale and determine the exact dimensions of low-order interaction spaces. We derive finite-sample risk bounds for harmonic projection and establish a nonasymptotic minimax characterization when the Johnson–Sobolev energy budget is controlled relative to the noise variance. We also show that stable least squares removes the signal-dependent fluctuation of empirical projection when the empirical Gram matrix is sufficiently well conditioned. For arbitrary signal-to-noise ratios, a rank-deficiency analysis quantifies the components left unidentified by the random design. Together, the rank-deficiency analysis and a centered completion estimator yield minimax bounds that match up to constants whenever \(\min(k,d-k)\) and the smoothness order are fixed. Numerical experiments reported in the Supplementary Material illustrate estimation under different interaction profiles and distinguish the effects of observation noise, random-design fluctuation, and incomplete coverage.
Read the full abstract
Combinatorics
A 3-regular counterexample to the Bilu–Linial signing conjecture
We construct a finite connected simple cubic graph $F$ such that every signing of its edges yields a signed adjacency matrix with an eigenvalue outside $[-2\sqrt2,2\sqrt2]$. This disproves the Bilu–Linial signing conjecture for general regular graphs. The graph $F$ is not Ramanujan, and the conjecture restricted to Ramanujan base graphs remains open.
Read the full abstract
Number Theory
An extension of algebraic independence of special values for non-lacunary power series
We study the algebraic independence of special values of power series $f(\beta^{-1})$, where $\beta$ is a fixed Pisot or Salem number. In particular, we consider the case where $f(X)=\sum_{n\geq 0} t(n) X^{w(n)}$ is not a lacunary series and is not assumed to satisfy any special functional equation, such as a Mahler-type functional equation. In our main results, we give a new criterion of the algebraic independence of three values. Applying our main results, we prove that the following three values are algebraically independent: \[ % \begin{gathered} \sum_{n=3}^{\infty}\lfloor n^{y}\rfloor\beta^{-\lfloor n^{\log \log n}\rfloor},\quad \sum_{n=3}^{\infty}\beta^{-\lfloor n^{\log \log n}\rfloor},\quad \sum_{n=1}^{\infty}\beta^{-\lfloor n^{\log n}\rfloor}, % \end{gathered} \] where $y$ is an arbitrary positive real number. Since our criterion is flexible, we have considerable freedom in choosing the coefficients $(t(n))_{n\geq 0}$ and the exponents $(w(n))_{n\geq 0}$.
Read the full abstract
Algebraic Geometry
Polarized varieties without arithmetically Cohen-Macaulay bundles and section rings without graded maximal Cohen-Macaulay modules in characteristic zero
We exhibit smooth polarized surfaces $(Y,H)$ over $\mathbf{C}$ carrying no nonzero arithmetically Cohen-Macaulay bundle of any rank. Equivalently, their section rings are three-dimensional normal $\mathbb{N}$-graded $\mathbf{C}$-domains, with an isolated singularity at the vertex, admitting no nonzero finitely generated graded maximal Cohen-Macaulay module. To the author's knowledge no such ring was previously known. Their existence contrasts with the theorem of Hartshorne, Hochster and Peskine-Szpiro that a three-dimensional $\mathbb{N}$-graded domain over a perfect field of characteristic $p>0$ always has one. As a consequence, in characteristic zero Hochster's small Cohen-Macaulay conjecture admits no graded refinement. The nonexistence criterion is numerical: for $|H|$ base-point-free with finite morphism, $H^2<K_Y^2-8\chi(\mathcal{O}_Y)=\tau(Y)$, the signature of $Y$ over $\mathbf{C}$, leaves no nonzero $\mathcal{E}$ with $H^1(Y,\mathcal{E}(tH))=0$ for all $t$. For Ulrich bundles, being semistable, this is Bogomolov's inequality; what is new is that no stability hypothesis and no condition on the Hilbert polynomial are needed, the Harder-Narasimhan filtration of $\mathcal{E}$ being compared instead with the Horrocks splitting of its direct image on $\mathbb{P}^2$. Hirzebruch's Hesse surfaces with $H_n=4A_n-E_n$, $n\ge3$, qualify. In dimension $m$ the obstruction reads $(m+1)H^m\ge(K_X^2-2c_2(X))\cdot H^{m-2}$, the inequality Lopez proved for Ulrich bundles; products and complete intersections inside them give examples in every dimension $\ge3$, and the threefold examples show that the characteristic-$p$ hypothesis in the criterion of Shimomoto-Tavanfar is essential.