Peer Revenue Past citations are no guarantee of future returns

Papers

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

All of arXiv

Saturday 12 September

Quantum Physics

Partial Stabilizer Learning under Fixed Commuting Constraints and the Absence of a Copy-Rate Discount

We consider an unknown pure stabilizer state of \(n\) qudits with a fixed prime local dimension \(p\), together with prescribed commuting Pauli observables. Rather than learning the whole state, the learner must recover only complementary stabilizers that commute with the prescribed observables, together with their eigenvalues. Once the prescribed measurement is performed, this information and the observed outcome determine the corresponding conditional stabilizer state. We compare the number \(k\) of copies available to the learner with the number \(n\) of qudits and refer to their ratio \(k/n\) as the copy rate. Suppose that the dimension \(m\) of the requested complementary stabilizer subspace satisfies \(m=\beta n+o(n)\) for a fixed \(0<\beta\leq1\). If the copy rate converges to a value below one, the optimal exact recovery probabilities and verification scores converge to zero, both in the worst case and under the uniform prior. Conversely, if the number \(k\) of copies minus the number \(n\) of qudits tends to positive infinity, the entire stabilizer state can be identified with probability tending to one, and the requested complementary information can then be extracted. Thus, restricting the learning target to the stabilizer information needed after the prescribed measurement yields no copy-rate discount.
Read the full abstract

2609.13923 · 12 Sep · A Lean 4 formalization of our main theorem, Theorem 2.2, is available. The formalization, verification records, and reproducibility instructions are publicly available in release v1.0.0 of the accompanying repository: https://github.com/Imine20/Lean-formalization-for-stabilizer-state-identification/releases/tag/v1.0.0

Machine Learning

Minibatch persistency, eight years later: what batch reuse costs in steps and joules, and what it saves in data

Minibatch persistency reuses data instead of reading it: rather than drawing a fresh minibatch at every optimizer step, it takes K consecutive steps on the same one. Absorbed into data echoing in 2019, it has carried one objection – that reuse merely imitates a larger learning rate – and no baseline tuned as carefully as the method itself. This paper runs the missing test. A pre-registered study trains a 49M-parameter Transformer on FineWeb-Edu at minibatch size B in {32, 128, 512}, 8 seeds per cell, tuning the learning rate separately for every batch size and every arm, against a reuse-free control that changes the sampling and nothing else. Each headline claim is a cost to reach a fixed loss, read on four axes: optimizer steps, fresh tokens, seconds, and joules at the socket. We then replicate on new seeds and a newer GPU generation, and put the three arms on one schedule in steps. The outcome of our study is that what minibatch reuse buys is neither speed nor energy but data, and only at large minibatch size: at B = 32 it reads more fresh tokens than the baseline, not fewer. On steps, seconds and joules it is at best free; and at B = 512, where it looks best, a registered control cannot separate the effect of reuse from the position on the learning-rate schedule at n = 8 seeds. The technique is therefore worth using where fresh data rather than compute is the binding cost: a corpus that runs out, a pipeline that pays per sample, a stream that cannot be rewound. Where the data can simply be read again, spaced epochs do as well or better.
Read the full abstract

Algebraic Geometry

Conifold Gap for the Quintic Threefold

We prove the all-genus conifold gap conjecture for the Gromov–Witten (GW) theories of the quintic threefold, other smooth Fermat Calabi–Yau threefold hypersurfaces, and local $\mathbb P^2$, as well as for the FJRW theories associated with these hypersurfaces. Our proof uses Mixed Spin P-field (MSP) theory and analogous master-space constructions. The central object is the cone-vertex theory. We express this theory as a translation of the point CohFT via a formal Laplace transform and prove a universal conifold gap theorem for the resulting translated theories. The master-space construction then transfers the gap from the cone-vertex theory to the GW and FJRW theories.
Read the full abstract

All authors Huai-Liang Chang, Shuai Guo, Lei You, Huaigong Zhang

Machine Learning

Finite-Time Node Separation in Recurrent Graph Neural Networks with Persistent Gaussian Perturbations

Persistent Gaussian perturbations have been shown to prevent asymptotic oversmoothing in recurrent Graph Neural Networks (GNNs) by ensuring a positive stationary Dirichlet energy. However, this global energy bound does not guarantee that individual node representations remain distinct at finite depths. In this paper, we provide a complementary finite-time analysis of the same persistent-noise architecture. Let \(d\) denote the representation dimension and \(\sigma\) the noise standard deviation. We first prove an exact second-moment decomposition for the expected squared distance between any two node representations, yielding the universal lower bound \(2\sigma^2 d\) at every positive time step without contraction or stationarity assumptions. More precisely, conditional pairwise distances have a noncentral chi-square characterization: the noncentrality parameter is the deterministic message-passing separation normalized by \(2\sigma^2\). This yields dynamics-aware fixed-time and finite-horizon near-collision bounds that retain information discarded by the central worst-case analysis. The earlier central Gaussian bound is recovered as the worst-case zero-separation case. We additionally prove almost-sure pairwise noncollision, derive a uniform finite-horizon guarantee, and establish permutation equivariance in distribution for the stochastic dynamics and permutation-invariant graph outputs. Our results complement the asymptotic energy analysis of prior work and provide rigorous finite-time guarantees on node-level representation separation.
Read the full abstract

Machine Learning

Machine Learning in Fish Farming

This chapter explores how machine learning (ML) is transforming aquaculture, with a particular focus on enhancing decision-making processes and improving operational efficiency. The chapter is structured to first introduce the challenges in aquaculture and the role of AI and then provide an overview of ML techniques in the context of aquaculture, followed by applications, emerging trends, future directions, and case studies. The focus is on real-world applications of ML techniques, including Random Forest, Convolutional Neural Networks (CNNs), and Recurrent Neural Networks (RNNs), as well as emerging technologies such as Graph Neural Networks (GNNs) and large language models (LLMs). Key applications include biomass estimation, species recognition, behavioural analysis, and environmental forecasting. The chapter also highlights the synergy between ML and the Internet of Things (IoT) for real-time monitoring and decision support. Ultimately, ML-driven innovations have the potential to revolutionise fish farming, leading to more efficient, sustainable, and productive practices in the aquaculture industry.
Read the full abstract

All authors Fearghal O'Donncha, Nikos Papandroulakis, Jennie Korus, Abigail Langbridge, Alexander Timms, Konstantinos Topouzelis, Abdul Baseer Khan, Shree Rama Kamal Kumar Vegu, Mahtab Sarvmaili, Ryan Mowat, Rhanna Turberville, Tyler Sclodnick, Christopher Whidden

Artificial Intelligence

LoRA Fine-Tuned Models for Control Systems Course Q&A: A Multidimensional Evaluation of Model Scale and Rank Effects

Large language models (LLMs) are increasingly used in specialized university courses, but control-systems questions require coordinated terminology, notation, derivations, and stepwise explanations. Direct general-purpose responses may be inconsistently structured and hard to verify. Using exercises and reference solutions from a Linear Control Systems course, we built a supervised fine-tuning dataset of 360 system-user-assistant conversations. We applied LoRA to Qwen2.5-3B-Instruct and Qwen2.5-7B-Instruct. With identical data splits, inference settings, and evaluation protocols, we compared base and fine-tuned models and tested LoRA ranks r=4, 8, and 16. Evaluation used ROUGE, BERTScore, and structured-output features to measure reference-answer similarity and stability of the Solution-Method-Teaching Points format. LoRA improved both similarity and structured-output stability at both sizes. On the current test set, 7B-r16 achieved the highest ROUGE-L (0.4093) and BERTScore-F1 (0.8643), while r=8 offered a better balance between performance and parameter efficiency. Bootstrap resampling showed ROUGE-L gains of 0.0764 [0.0613, 0.0915] for 3B-r16 and 0.0874 [0.0687, 0.1042] for 7B-r16; both intervals exceeded zero, indicating stable textual-similarity improvements on the current test set. These results suggest LoRA can align open-source instruction-tuned models more closely with the language and pedagogical organization of course reference answers. However, the metrics mainly capture textual similarity and formatting consistency, not domain-specific reasoning or mathematical correctness, which require expert assessment and task-specific rubrics.
Read the full abstract

All authors Shaowen Lu, Chengxu Liu, Ping Zhou, Tao Yang

Machine Learning

Learning Through Energy Refinement and Manifold Projection: A Cooperative EBM-AE Framework

Energy-Based Models (EBMs) provide a flexible framework for generative modeling by learning an energy landscape that assigns low energy values to realistic samples and higher energies to unlikely observations. Despite their theoretical appeal, training EBMs remains challenging due to the computational cost of Langevin sampling and the difficulty of efficiently exploring the learned data manifold. In this work, we propose a cooperative Energy-Based Model and Autoencoder (EBM-AE) framework that combines energy-based refinement with manifold projection. The proposed approach jointly trains an EBM with a denoising autoencoder and introduces an iterative EBM$\rightarrow$AE$\rightarrow$EBM sampling procedure in which Langevin dynamics and autoencoder projection alternately refine generated samples. Within this framework, the autoencoder acts as a manifold projection operator that regularizes sampling trajectories, while the EBM performs energy-based refinement toward low-energy regions of the learned distribution. Extensive experiments conducted on the MNIST dataset demonstrate that joint EBM-AE training substantially improves generation quality compared with a conventional autoencoder. Beyond unconditional generation, we evaluate the proposed framework on image inpainting tasks involving structured and random masks. The results show that manifold projection provides the majority of the reconstruction capability, whereas the final energy-based refinement becomes increasingly beneficial as the reconstruction problem becomes more challenging. Taken together, the results indicate that combining manifold projection and energy minimization provides an effective and interpretable framework for generation, reconstruction, and out-of-distribution detection, while offering new insights into the complementary roles of energy-based modeling and representation learning.
Read the full abstract

Computation and Language

North Small Translate: Advanced Cost-Effective Translation (Cohere CAT+)

We present North Small Translate, an open-weight, LLM-based machine translation (MT) model with instruction-following capabilities built on the same foundation as Cohere's Command A Plus, a mixture-of-experts architecture with 25 billion active parameters out of 218 billion total parameters. North Small Translate is trained using difficulty sampling to obtain challenging documents and a five-step training protocol combining supervised fine-tuning, direct preference optimization, and online reinforcement learning. We prioritized throughput through a non-reasoning base model and supplemented with optional agentic capabilities to unlock translation quality gains. North Small Translate is trained to perform MT-related tasks, including post-editing and quality estimation, as well as related tasks such as general instruction following. The model achieves top MT performance across 50 languages in the class of models under 1T parameters, with no need to run expensive reasoning at inference time.
Read the full abstract

All authors Tom Kocmi, Alexandre Bérard, Phil Blunsom, Samuel Cahyawijaya, Shaun Cassini, Nicholas Frosst, Ona de Gibert, Aidan Gomez, Nithya Govindarajan, Shun Kiyono, Olivia Lasche, Lawrence Rogers, Kelly Marchisio, Nikita Moghe, Yash More, Camila Moran-Hidalgo, Yiyang Nan, Michael Sachs, Trisha Starostina, Daan van Stigt, Spencer Rarrick, Sebastian Vincent, Ivan Zhang

High Energy Physics (Phenomenology)

Invariant Algebras of Low-Rank Neutrino Matter Flows: Krylov-Plücker Invariants, the $3+1$ Toric Ring, and Controlled Deformations

Matter changes the eigenvalues and mixing matrix of the neutrino propagation Hamiltonian while entering the flavor-basis Hamiltonian through a low-dimensional additive deformation. We develop a unified algebraic description of the quantities that are insensitive to this deformation. The invariant algebra is formulated as the joint kernel of commuting matter-translation derivations inside the flavor-rephasing invariant ring. For diagonal matter potentials, this gives a general decomposition into unaffected diagonal directions and an off-diagonal cycle algebra, together with an exact dimension formula for an arbitrary number of flavors and independent matter parameters. For matter spurions with common low-rank support, block-Krylov rows generate exterior-power covariants whose maximal minors are matter independent. Their mass-basis form is a Plücker expansion: rank one produces the familiar Vandermonde factorization, whereas higher rank generically produces a sum of independent Plücker terms. The standard $3+1$ system provides our principal nontrivial example. Its global off-diagonal algebra is the toric cycle ring of the complete four-vertex flavor graph, generated by edge moduli, triangle cycles, and quadrangle cycles; the commonly used eleven invariants constitute a generic local coordinate system rather than a global polynomial generating set. We further formulate independently varying matter compositions and off-diagonal nonstandard interactions as, respectively, commuting exact flows and controlled deformations of the invariant algebra. This framework separates the existence of exact invariants from the stronger and exceptional property of single-monomial spectral factorization.
Read the full abstract

Machine Learning

Machine Learning under Imperfect Data: Challenges and Methods

Machine-learning models are commonly developed under an assumption that training and test data are sufficiently complete, balanced, labelled, and drawn from compatible distributions. In practice, one or more of these conditions is often violated. Measurements may be missing or corrupted, rare classes may be poorly represented, supervision may be weak, and the deployment environment may differ from the training environment. These imperfections are usually treated as separate technical problems, although they alter learning through a small number of shared mechanisms: loss of information, biased empirical risk, ambiguous supervision, and unstable representations. This short survey organises representative methods around these mechanisms. It reviews reconstruction and generation, rebalancing and representation calibration, learning with limited supervision, adaptation across domains and modalities, and reliability under distribution change. The discussion highlights the limits of plausible reconstruction, benchmark-specific correction, and adaptation without trustworthy feedback. It concludes with directions for evidence-aware learning, uncertainty-preserving prediction, and evaluation that separates visual plausibility from decision utility.
Read the full abstract

Nuclear Experiment

Low-energy collective structure of $^{92,94}$Zr and $^{94}$Mo from ($e,e^{\prime}$) and ($p,p^{\prime}$) scattering II. Signatures of mixed-symmetry states and origin of collectivity

This is the second of two papers discussing a new experimental signature of mixed-symmetry states (MSS) in the vibrational nuclei $^{92,94}$Zr and $^{94}$Mo based on proton and neutron transition densities. Quasiparticle-phonon model calculations for these nuclei are extensively tested by comparison to ground-state properties, energies and moments of excited states, transition probabilities between them, and the momentum transfer dependence in $(e,e^\prime)$ and $(p,p^\prime)$ reactions. The overall very good agreement permits the extraction of information on the wave functions of the MSS and their fully symmetric (FSS) counterparts. MSS can be identified by a sign change between the leading proton and neutron two-quasiparticle configurations compared to the FSS. Possible candidates for $2^+$, $3^-$, and $4^+$ MSS are identified. The modification of proton and neutron transition densities, which can be derived from a combined analysis of inelastic electron and proton scattering, by the sign change provide a new experimental signature of MSS. The collectivity of ground-state excitations of predominantly one-phonon FSS and MSS is generated to a large extent by the coupling to high-lying states forming giant resonances with the same spin and parity.
Read the full abstract

All authors C. Walz, L.M. Donaldson, P. von Neumann-Cosel, N. Pietralla, F.D. Smit

Materials Science

Unbiased first-principles construction of complete tensorial spin Hamiltonians

Magnetic ground states are commonly predicted using spin Hamiltonians whose interaction terms are selected a priori, potentially overlooking the microscopic interactions that govern complex magnetic order. Here, we introduce a general framework for the unbiased first-principles construction of symmetry-complete tensorial spin Hamiltonians and its automated implementation in AMATIS. The framework constructs the Hamiltonian directly from density-functional theory while rigorously enforcing quantum spin algebra and crystallographic symmetry. Applied to representative two-dimensional van der Waals magnets, the framework reproduces established magnetic interactions and uncovers hidden physics beyond conventional spin models, including chiral interactions that stabilize metastable skyrmions, higher-rank tensorial interactions that reconstruct the magnetic phase diagram and establish stabilizing competing multi-Q phases, and an emergent p-wave altermagnetic electronic structure. Our results demonstrate that unbiased tensorial Hamiltonian construction provides a predictive alternative to the conventional practice of manually selecting spin-model interactions, enabling first-principles discovery of unconventional magnetic phases.
Read the full abstract

All authors Haichang Lu, Boyang Deng, Hiroshi Katsumoto, John Robertson, Weisheng Zhao, Stefan Blügel

Audio and Speech Processing

DualSpecSE: A Dual-Path Speech Enhancement Network Integrating Mel and Complex Spectrograms

In this paper, we propose DualSpecSE, a speech enhancement framework that jointly models Mel-spectrogram and complex spectrogram in a dual-path architecture for improved ASR performance and higher-quality speech reconstruction. The Mel branch learns coarse-grained acoustic representations and produces enhanced Mel-spectrograms for direct ASR usage, while the complex branch refines fine-grained spectral details for high-fidelity waveform reconstruction. Built upon the cross-band and narrow-band blocks from CleanMel, DualSpecSE introduces an interaction module and a fusion module to enable effective information exchange between the two branches. The model simultaneously outputs enhanced Mel and complex spectrogram without requiring a pretrained vocoder. Experimental results demonstrate consistent improvements in speech fidelity, perceptual quality, and ASR performance. Codes and audio samples are available.
Read the full abstract

All authors Xingchen Li, Ziqian Wang, Zikai Liu, Yike Zhu, Zihan Zhang, Longshuai Xiao, Lei Xie

Materials Science

Quantum interpretation of classical LiF simulations

Recent studies have shown that classical LiF simulations can reproduce low-temperature experimental spectra when performed at a simulation temperature $T^*$ higher than the physical temperature $T$. We show that the characteristic scale of this rescaling follows naturally from quantum vibrational statistics. An effective temperature $T_{\rm eff}(T)$ is obtained by matching classical and quantum harmonic energies and is evaluated using an analytic acoustic-optical model and a first-principles LiF vibrational density of states $D(\varepsilon)$. The resulting temperature reproduces the low-temperature scale of the reported spectral temperature and approaches the physical temperature in the classical regime. Our results identify zero-point motion and Planck-Bose-Einstein statistics as the origin of the reported temperature scale.
Read the full abstract

Sound

DiTAR+: Dual Optimization for Robust Autoregressive Diffusion Speech Synthesis

Continuous-latent Autoregressive Diffusion Transformer (AR-DiT) models have demonstrated immense potential in zero-shot speech generation. However, they still suffer from limited decoding stability when synthesizing long utterances or complex linguistic structures. This instability primarily stems from a restricted historical receptive field and an acoustic inertia dependency within the diffusion decoder, which causes the model to ignore semantic conditions. To address these challenges, we propose DiTAR+, a dual-optimization framework. First, we introduce Dilated Context Sampling to expand the macro-level historical receptive field without violating physical temporal continuity, thereby preventing cumulative error propagation. Second, we propose Hierarchical Acoustic Masking to prevent shallow layers from attending to acoustic pre-context, explicitly decoupling semantic alignment from acoustic detail reconstruction. Extensive experiments show that our framework effectively mitigates pronunciation errors and semantic hallucinations, enhances generation robustness on challenging sentences, and maintains exceptionally high speaker similarity throughout the entirety of long-form utterances. On the linguistically challenging ZH-Hard set, DiTAR+ reduces the word error rate from 12.478% to 9.893%, and on extended utterances of 25 to 35 seconds it improves speaker similarity from 0.741 to 0.759 while simultaneously lowering the word error rate from 2.778% to 2.173%, outperforming both discrete-token and pure flow-matching baselines.
Read the full abstract

All authors Ziyu Zhang, Tianlun Zuo, Hanzhao Li, Haoyu Zhang, Lei Xie

K-Theory and Homology

The Nil K-groups of finite groups

We show that for every finite group $G$ and every integer $n$, the Nil group $\mathrm{NK}_n(\mathbb{Z}[G])$ of the integral group ring of $G$ has finite exponent, and give a bound depending on n and the order $|G|$. This bound is explicit when every prime divisor of $|G|$ is at least $3+n/2$. More generally, our finite exponent result applies to $\mathrm{NK}_n(R[G])$ whenever $R$ is a regular, torsion-free, Noetherian commutative ring such that $R/p$ is regular for every prime $p$ that divides $|G|$. In the Appendix, M. Morrow provides an alternative approach and also shows that the Nil group $\mathrm{NK}_n(X)$ of an excellent Noetherian scheme $X$, with $X[1/p]$ regular, is annihilated by a finite power of $p$, provided $X$ admits a suitable resolution of singularities. By extending his argument to certain noncommutative rings, we also show that, for $R$ as above and $\alpha$ any automorphism of $G$, the Farrell Nil groups $\mathrm{NK}_n(R[G],\alpha)$ have finite exponent. As a consequence, for every virtually cyclic group $\Gamma$, the group $\mathrm{K}_n(\mathbb{Z}[\Gamma])$ is a direct sum of a finitely generated abelian group and an infinite countable direct sum of copies of a finite abelian group.
Read the full abstract

All authors Ted Chinburg, Matthew Morrow, Georgios Pappas, Martin J. Taylor

Computation and Language

Phorecaster365: A Human-Supervised Reference Architecture for Hybrid Pharmaceutical Sales Forecasting and Planning Decision Support

Pharmaceutical sales forecasts inform planning across products, regions, and distribution channels, yet their interpretation depends on inventory availability, transaction semantics, product lifecycle, and the information available when each forecast is issued. A model prediction alone does not preserve these conditions or establish whether a forecast is suitable for operational use. We present Phorecaster365, a human-supervised reference architecture that connects enterprise resource planning data to reviewable pharmaceutical sales forecasts. The architecture separates source ingestion, product-region series construction, temporally valid feature generation, statistical and machine-learning modeling, ensemble formation, uncertainty assessment, planner review, and lifecycle governance. Its central intermediate representation is a forecast context package that preserves the source snapshot, forecast target, temporal cutoff, available covariates, data-quality state, and hierarchy version. A corresponding forecast evidence package links predictions to model and calibration versions, exceptions, human adjustments, and publication history. Development experience with a synthetic panel of 10,950 daily records across 30 product-region series informs the design. Historical experiment summaries are retained only as descriptive evidence of development because their evaluation does not establish independent predictive validity. We specify a rolling-origin evaluation protocol, baseline and ablation requirements, uncertainty and robustness assessments, and a staged pathway from synthetic testing to use in governed planning. The contribution is an implementation-neutral system design and validation framework; the report does not establish real-world forecasting accuracy, comparative superiority, or operational benefit.
Read the full abstract

Functional Analysis

Failure of Rockafellar's Sum Conjecture via an Explicit Non-(FPV) Operator

Building on a construction of Weifeng Yang, we present a simplified counterexample to Rockafellar's sum conjecture on $c_0(\N_0\times\N)$. The central observation is geometric: the domain of a maximally monotone operator can have a nonconvex norm closure. In our example, this gives a short proof that the operator is not of type (FPV), which in turn implies the failure of the sum conjecture. We define an explicit operator using geometric series and provide a detailed proof of its maximal monotonicity. The example was developed with assistance from GPT-6 Astra. The purpose is to simplify and explain the construction, not to claim an independent counterexample mechanism.
Read the full abstract

Combinatorics

Zeros in The Character Table of The Wreath Product of The Symmetric Group

Let $G$ be a finite group with $t$ conjugacy classes, and let $S_N$ be the symmetric group. Let $Z_t(N)$ be the number of zeros in the character table of the wreath product $G\wr S_N$. We prove \begin{equation*} Z_t(N)\ge \frac{2p_t(N)^{2}}{\log \frac{N}{t}}\left(1+O\left(\frac{\log \log \frac{N}{t}}{\log \frac{N}{t}}\right)\right), \end{equation*} where $p_t(N)$ is the number of $t$-multipartitions of $N$.
Read the full abstract

Number Theory

Generalized Wieferich primes and monogenic polynomials

Let $b, p\in {\mathbb Z}$ with $b\ge 2$ and $p\ge 3$ a prime. If $b^{p-1}\equiv 1 \pmod{p^2}$, then $p$ is called a {\em generalized Wieferich prime base $b$}, or more succinctly, a {\em base-$b$ Wieferich prime}. When $b=2$, $p$ is also known simply as a Wieferich prime. We say that a monic polynomial $f(x)\in {\mathbb Z}[x]$ is monogenic if $f(x)$ is irreducible over ${\mathbb Q}$ and $\{1,\theta,\theta^2,\ldots,\theta^{\deg(f)-1}\}$ is a basis for the ring of integers of ${\mathbb Q}(\theta)$, where $f(\theta)=0$. Recently, necessary and sufficient conditions for the monogenicity of the trinomials $x^{2n}+bx^n+b$ were given that included base-$b$ Wieferich prime congruences, and also congruences involving a certain Lucas sequence. In this article, we prove a similar result for a different class of trinomials that does not rely on any congruence condition involving a Lucas sequence. Furthermore, we extend this result to a related class of $N$-nomials, for any $N\ge 4$.
Read the full abstract

Computer Vision

SkyAnchor: Updating Metric-scale Aerial 3D Gaussian Scenes from Unposed Ground-View Sequences

We study how to update a pre-built aerial scene with a newly captured, unposed ground-view sequence. The aerial scene already contains a reliable metric Structure-from-Motion (SfM) reconstruction and a pre-trained 3D Gaussian Splatting (3DGS) model, whereas the ground-view sequence is collected later to add street-level appearance but has unknown camera poses and global scale. Registering this sequence to the aerial SfM reconstruction is challenging because single-image cross-view localization is brittle and long trajectories are prone to drift. To address these challenges, we present SkyAnchor, which treats the existing aerial scene as a fixed scaffold for ground-view registration and scene update instead of jointly reconstructing aerial and ground imagery from scratch. It first localizes short groups of consecutive ground frames against geometrically verified aerial support, producing sparse anchor poses. It then recovers the full ground trajectory with anchor-constrained submaps, fixing the front and rear anchor poses during incremental registration and bundle adjustment. Finally, it inserts filtered ground Gaussians while preserving the aerial view, followed by lightweight joint refinement. Experiments on seven real aerial–ground scenes show accurate metric ground trajectories and updated 3D Gaussian scenes with strong aerial- and ground-view rendering quality.
Read the full abstract

All authors Zhuoxiao Li, Xinyi Liu, Taoyu Wu, Yinrui Ren, Tongyan Hua, Ou Jing, Shuai Zhang, Dongli Wu, Rongjun Qin, Ge Lin Kan, Wufan Zhao

Nuclear Experiment

Low-energy collective structure of $^{92,94}$Zr and $^{94}$Mo from ($e,e^{\prime}$) and ($p,p^{\prime}$) scattering I. Experimental results

This is the first of two papers discussing a new signature of quadrupole mixed-symmetry states in the vibrational nuclei $^{92,94}$Zr and $^{94}$Mo based on proton and neutron transition densities derived from the comparison of inelastic electron and proton scattering experiments. Results of the $(e,e^\prime)$ and $(p,p^\prime)$ experiments on these nuclei as well as $(p,p^\prime)$ data for other potential cases $^{96}$Mo and $^{70}$Zn are presented. Spin and parity quantum numbers of excited states are assigned based on angular distributions from the proton scattering data in comparison to calculations employing form factors from the collective model in distorted-wave Born approximation. Possible candidates of one-phonon fully symmetric and mixed-symmetry states as well as members of two-phonon multiplets are identified.
Read the full abstract

All authors C. Walz, O. Burda, L.M. Donaldson, R.W. Fearick, S.V. Förtsch, H. Fujita, P. von Neumann-Cosel, R. Neveling, N. Pietralla, A. Richter, F.D. Smit, C. Stahl, J.A. Swartz

Networking

HQARRF: Hierarchical Q-learning and Force-aware Routing for Multi-Charger Scheduling in Wireless Rechargeable Sensor Networks

Multi-charger scheduling in wireless rechargeable sensor networks must weigh sensor death risk, charger energy, travel cost, return-to-base feasibility and inter-charger coordination at once, and schedulers driven by local urgency alone duplicate service and leave whole regions unattended. We present HQARRF, a two-level scheduler. Below, an interpretable ARR-F score ranks candidate clusters through an attraction term for local urgency, a repulsion term against charger crowding and a force bonus from nearby critical sensors. Above, adaptive zones compress regional state into a deadline-based risk estimate, and a gated Q-learning controller decides only whether to redirect service to a high-risk, under-served zone. Over 27 parameter points HQARRF attains the highest mean survival rate at 26, improving survival by 20.7 percentage points over the mean of five baselines and 9.2 over the strongest baseline at each point. An ablation isolates the upper level: its gain tracks how often the controller fires.
Read the full abstract

Combinatorics

On additive complements in natural numbers

We prove that for additive complements $A$ and $B$, if $$ \limsup_{x\to+\infty}\frac{A(x)B(x)}{x}<\frac43, $$ then $$ A(x)B(x)-x\rightarrow+\infty\qquad(x\to+\infty). $$ This improves the earlier upper bound $3-\sqrt{3}$ due to Fang and Chen.

Neurons and Cognition

URCHIN: A Horizontal Spiking Language Model for Data-Constrained Pretraining

The BabyLM challenge measures how much language a model can learn from developmentally-plausible, child-scale data rather than internet-scale corpora, yet prior language models forgo the biological constraints of the neural circuitry that acquires human language: spiking neurons separated into excitatory and inhibitory populations wired by a recurrent lateral connectome. This paper presents URCHIN (Unified Recurrent Connectome with Horizontal Integrate-and-fire Neurons), which applies the Parallelized Hierarchical Connectome Spiking State-space Model (PHCSSM) to language modeling: leaky integrate-and-fire neurons coupled by a Dale's-law lateral connectome resolve each token through a multi-transmission loop that recirculates activity to a fixed point. The instantiation is deliberately minimal: a single horizontal layer of 128 neurons, no attention, and 4.23M parameters. Two implementations share one set of weights and produce identical benchmark scores, so URCHIN is trained once and deployed either way with no conversion step: a parallel state-space model (SSM) scan that is GPU-efficient for training, or an event-driven recurrent spiking neural network (RSNN) with constant-cost inference for CPU or neuromorphic edge deployment. Across all three BabyLM tracks (Strict-100M, Strict-Small, and Multilingual), URCHIN offers a biologically plausible, efficient, and directly deployable reference point.
Read the full abstract

Statistics Theory

Risk comparison theorems and application to deep learning of diffusion coefficients

We investigate the nonparametric estimation of the diffusion matrix in stochastic differential equations featuring multidimensional, strong mixing covariate processes. We propose a flexible statistical framework based on general function classes that does not require a linear basis representation, rendering our results directly applicable to deep neural network estimators. Our approach employs a two-step estimation procedure: constructing a preliminary nonparametric quasi-likelihood estimator and subsequently regularizing it via a $\beta$-Hölder class approximation. We establish general risk comparison theorems between empirical and generalization risks for arbitrary estimators without relying on a specific probabilistic structure of the underlying process. In diffusion matrix learning based on $n + 1$ observations over the time interval $[0,T]$, the derived upper bounds capture the intrinsic interplay between the $T$-rate, associated with the mixing behavior of the covariate process, and the intrinsic $n$-rate governing the volatility estimation.
Read the full abstract

General Relativity and Quantum Cosmology

A new parametric study of $f(T)$ teleparallel gravity with generalized Chaplygin gas: Confronting observational data

In this paper, we investigate the cosmological dynamics of a modified gravity framework based on the function $f(T)=\beta(-T)^{1/2}+\gamma(-T)$, where $T$ denotes the torsion scalar. The matter sector is modeled using the Generalized Chaplygin Gas (GCG) with the equation of state $p=-\frac{\mathcal{A}}{\rho^{\eta}}$, allowing the model to describe the evolution from an early-time matter-like phase to a late-time accelerated universe. By deriving an analytical expression for the Hubble parameter $H(z)$, we perform parameter estimation using Bayesian statistical techniques based on the $\chi^{2}$-minimization method with the Cosmic Chronometer (CC) and joint (CC+Pantheon) observational datasets. The deceleration parameter exhibits a transition from deceleration to acceleration, while the present-day value indicates the current accelerated expansion. The model yields a present-day EoS parameter consistent with dark-energy-like behavior, while the energy conditions indicate that the NEC and DEC are satisfied and the SEC is violated at late times. The $\omega-\omega'$ plane shows a freezing quintessence behavior, approaching the $\Lambda$CDM-like point $(\omega,\omega')=(-1,0)$, while the cosmographic analysis offers additional insight into the evolution of the cosmic expansion. The estimated age of the universe is also consistent with current observational bounds. These results demonstrate that the GCG scenario within $f(T)$ teleparallel gravity provides a viable and observationally consistent framework for explaining the late-time accelerated expansion of the universe.
Read the full abstract

Optics

Refractive Index Dispersion Fingerprinting via Scanning-Free Parallel Multi-Wavelength SPR on a Single Aluminum Film

Real-time characterization of refractive index (RI) dispersion is pivotal for advanced optical sensing, yet conventional surface plasmon resonance (SPR) platforms are bottlenecked by the narrow bandwidth of noble metals (Au, Ag) and the mechanical instability of sequential scanning. Here, we report a novel Al-based parallel SPR platform that overcomes the bandwidth and temporal constraints of conventional noble-metal systems. Leveraging the unique low-loss broadband response of Al, enabled by the suppression of interband transitions, we engineered a system for simultaneous excitation at 450, 520, and 635 nm. By integrating spectral-angle multiplexing with RGB-channel demultiplexing on a CMOS camera, the platform achieved acquisition of dispersion profiles without mechanical motion. Validated against NaCl solutions, the system demonstrates metrological accuracy and exceptional agreement with Cauchy dispersion models. The proposed architecture eliminates temporal drift and vibration errors, establishing a new benchmark for real-time dispersion characterization. By decoupling sensing from mechanical constraints, this work pioneers a compact, robust framework for next-generation, field-deployable sensors capable of distinguishing complex analytes via their unique spectral signatures.
Read the full abstract

All authors Zihao Luo, Zhiying Chen, Yueqian Zhang, Xue Liu, Changsen Sun, Dmitry Kiesewetter, Sergey Krivosheev, Sergey Magazinov, Victor Malyugin, Xue Han

Systems and Control

Influence Enhancement in Opinion Dynamics Using Edge Modification: A Kron Reduction-Based Approach

With the emergence of online social networks as a primary platform for advertising and advocacy, enhancing a user's influence has become of significant interest. In this paper, we investigate this problem under the Friedkin Johnsen opinion dynamics model, wherein stubborn agents influence the opinions of other agents in the network. Unlike most of the existing works, we leverage topological properties of the network to increase the influence of a desired stubborn agent. Specifically, we introduce the notion of edge modification, which mimics the mechanism of recommendations in social networks. First, we present a topology-based condition that identifies edge modifications that always increase the influence of a desired stubborn agent. It is shown that the impact of the chosen edge modifications remains robust to changes in parameters such as stubbornness and the interaction weights. Thereafter, we formulate a discrete optimisation problem to identify a set of edge modifications that maximise the agent's influence centrality. We present a computationally efficient approximate solution to the optimisation problem. Finally, we demonstrate the effectiveness of our approach on the Friedkin-Johnsen opinion dynamics over the Erdos Renyi random graph.
Read the full abstract

All authors Aashi Shrinate, Aravind Seshadri, Twinkle Tripathy, Laxmidhar Behera, Lingfei Wang, Karl Henrik Johansson

Number Theory

The optimal scale in Linnik's microsquare problem for almost all integers

We prove that, for any positive function $\Psi(t)\to\infty$, almost all positive integers $n$ satisfying $4\nmid n$ and $n\not\equiv7\pmod8$ can be written as $n=x^2+y^2+z^2$, where $x,y,z\in\mathbb{Z}$ and $1\leqslant z<\sqrt{\log n}\,\Psi(n)$. This improves Wooley's bound, and we show that the result is best possible. Our approach combines Hooley's weighted representation function for sums of two squares with Green's $W$-trick. We also give a necessary and sufficient condition, in terms of covering systems, for almost all odd integers to be represented as the sum of a squarefree number and a power of two, answering a question of Granville and Soundararajan.
Read the full abstract