Peer Revenue Returns subject to peer review

Papers

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

All of arXiv

Wednesday 2 September

Computer Vision

Structured-Prior-Guided Diffusion Inpainting with Physical Consistency for Traffic Sign Augmentation

Traffic sign detection faces a long-tailed data distribution. Many rare signs matter as much as common ones from a regulatory standpoint, yet they have very few samples. Generative data augmentation is one way out. General-purpose inpainting models, however, distort digits, deform geometry and perspective, and shift colours when applied directly to sign regions. We trace this to a single gap: the conditioning signal is too abstract for the physical composition of a sign. We propose a structured-prior-guided diffusion inpainting framework with physical consistency. It injects the semantic, appearance and geometric priors of a sign through three orthogonal pathways: a JSON-formatted text prompt, a front-view vector template rendered with measured dominant colours (via IP-Adapter), and an affine-aligned vector template (via ControlNet). Two physical consistency losses constrain colour with a CIELAB chromaticity $L_1$ term and edge structure with a Sobel gradient term. We train by self-supervised reconstruction on a large set of images collected in-house at AMAP, then evaluate zero-shot on the public TT100K-2021 dataset, a different source. Our method uses a Stable Diffusion 1.5 backbone of about 1.4B parameters. It beats seven representative competitors on every metric of reconstruction fidelity, physical consistency and semantic controllability. Its OCR exact-match rate reaches 91.1%, against 44.2% for the 12B industrial model FLUX.1 Fill [dev], and it needs only $1/14$ of that model's inference time. Leave-one-out ablations confirm that each of the three prior pathways and both loss terms contribute on their own. In downstream detection, the synthetic data raises the group-pooled AP50 of rare classes by $1.23\times$ to $7.40\times$ over a real-data-only baseline. Code and pre-trained models are available at https://github.com/52hz-whale/TrafficSignInpaint.
Read the full abstract

All authors Luo Li, Chongchong Huang, Jun Jia, Qiang Gao, Xinlong Liu, Gui Yang, Liang Cao

Numerical Analysis

A Stream-Function Formulation for the Surface Stokes-Helfrich Model - Surface Finite Element Discretization and Validation

We introduce a stream-function formulation for the Surface Stokes-Helfrich model and study an Arbitrary Lagrangian-Eulerian (ALE) Surface Finite Element (SFEM) discretization in space and a semi-implicit discretization in time of these equations. In contrast with special cases of a surface Stokes model on stationary or prescribed evolving surfaces the surface pressure remains as an unknown and needs to be reconstructed in each time step. As for the surface Stokes model on a prescribed evolving surface the formulation also requires an additional gradient potential. A detailed computational comparison with an ALE-SFEM discretization in space and a similar semi-implicit discretization in time of the corresponding velocity-pressure formulation is explored on a representative problem formulation for simply-connected surfaces, demonstrating the validity of the approach.
Read the full abstract

Number Theory

Regularised double shuffle relations for planar Arborified Zeta Values

We endow spaces of decorated planar rooted trees with new dendriform and tridendrifrom algebra structures and provide their combinatorial description. We then show that the planar counterparts of Arborified Zeta Values are algebra morphisms for these shuffle and quasi-shuffle products of planar rooted trees. We also prove an arborified version of Hoffman's regularisation relation for Arborified Zeta Values. We conjecture that those give every rational relation between Arborified Zeta Values and show that this conjecture implies the regularised double shuffle conjecture for Multiple Zeta Values.
Read the full abstract

Neurons and Cognition

Fungal Memory and Minimal Cognition

This paper argues that fungal mycelial networks exhibit minimal cognition through memory-integrated adaptive regulation. Drawing on cybernetic and enactivist frameworks, I develop a non-representational account of memory as the organism's capacity to modulate behavior based on temporally extended environmental coupling. I propose four operational criteria for minimal cognition: feedback-guided regulation of behavior, maintenance of internal viability conditions, structural modulation based on past environmental interactions, and plasticity across time scales that supports anticipatory adaptivity. Empirical evidence demonstrates that fungi meet all four criteria through distributed memory mechanisms: fungal networks exhibit directional regrowth toward previously encountered resources even after spatial displacement; stress priming persists across multiple cell divisions; Spitzenkörper-mediated directional persistence in constrained environments; and transgenerational memory through spore imprinting. These findings challenge representationalist assumptions in cognitive science by showing that memory and cognition can emerge from morphodynamic, biochemical, and electrophysiological processes without necessary neural substrates or symbolic representations. Fungal cognition demonstrates that the organizational principles underlying cognition - feedback-driven adaptation, norm-preservation, and historical coupling - can be realized in radically different material substrates, expanding our understanding of what counts as a cognitive system.
Read the full abstract

Genomics

Subcellularly Resolved Single-Cell Embedding Learning with Transcriptomic data, Protein Structure and Localization Information

Existing cell embedding methods predominantly rely on transcriptomic or proteomic measurements and represent each cell as a holistic entity, thereby overlooking the subcellular localization of individual molecules. Moreover, they rarely incorporate protein structural information, despite its fundamental role in determining molecular interactions and functions. In this work, we propose a multimodal framework for learning subcellularly resolved cell embeddings by jointly leveraging RNA expression profiles, protein sequence representations, and protein structural information. Specifically, we employ a cross-attention architecture to integrate transcriptomic, sequence, and structural modalities and model their interactions within distinct subcellular compartments. The resulting embeddings represent each cell through its fine-grained subcellular organization, capturing both molecular expression patterns and the functional properties of the associated proteins. By learning cell representations at subcellular resolution, our framework preserves spatially organized biological information while integrating complementary signals across multiple molecular levels. To the best of our knowledge, this is the first framework that produces subcellularly resolved cell embeddings by jointly incorporating transcriptomic information, protein sequence representations, and protein structural knowledge within a unified cross-modal learning paradigm.
Read the full abstract

All authors Zhen Zhou, Jiachen Li, Yuan Liu, Xiaoyong Pan, Hong-Bin Shen

Sound

SonicCaps: Large-Scale Diverse and Fine-Grained Captioning for Improved Audio-Retrieval

Recent advances in audio-language modeling have been driven by large-scale audio captioning datasets. However, existing datasets remain limited by low semantic diversity, generic descriptions lacking acoustic details, and one-to-one audio-caption mappings that poorly reflect the inherent ambiguity of auditory perception. We introduce SonicCaps, a large-scale audio captioning dataset comprising ~15M captions paired with ~700k audio clips, generated using a multi-modal large language model (Qwen3-Omni) conditioned on both audio and text. To explicitly promote diversity, we generate around 24 captions per audio via structured prompt engineering and few- shot generation, spanning main descriptions, rephrased variants (verbosity, style) and semantic tags. Human evaluation shows that SonicCaps is rated significantly higher than existing captioning datasets, with fine-grained analyses indicating that our captions are perceived as more descriptive and precise, which strongly correlates with quality judgments. Finally, training CLAP models on SonicCaps with a multi-caption sampling strategy consistently improves audio retrieval and zero-shot classification, with stronger generalization across public and commercial benchmarks. We release both SonicCaps and two specialized CLAP models on hugging face: https://huggingface.co/datasets/Zineb/SonicCaps.
Read the full abstract

All authors Zineb Lahrichi, Marc Ferras, Gaël Richard, Geoffroy Peeters

Functional Analysis

$\mathcal{F}$-Transitivity of Translation Semigroups on Directed Metric Trees

We study the $\mathcal F$-transitivity and topological $\mathcal F$-recurrence of left translation semigroups on weighted $L^p$-spaces over directed metric trees. Motivated by the recent work of Mangino and Vargas-Moreno on hypercyclicity and weak mixing for these semigroups, we investigate the different problem of $\mathcal F$-transitivity, where the entire return-time set is required to belong to a prescribed finitely invariant Furstenberg family. Assuming that the weight is $p$-admissible, we obtain necessary and sufficient integral conditions for both rooted and rootless trees, and establish the equivalence between $\mathcal F$-transitivity and topological $\mathcal F$-recurrence. In the rooted case, the criteria depend only on the weights along descendant edges. In the rootless case, an additional ancestor term and a measurable cancellation construction reflect the backward geometry of the tree. Our approach is based on measurable weighted minimization and quantitative estimates over subsets of the edge parameter interval. Examples on homogeneous rooted trees and rootless trees with a free left end show that branching may determine the dynamics, but forward branching alone does not guarantee $\mathcal F$-transitivity in the rootless setting.
Read the full abstract

Computer Vision

Towards Zero-Shot Transfer Across Embodiments For Driving VLAs

Vision-Language-Action models (VLAs) have shown strong potential in autonomous driving by leveraging multimodal pretraining for instruction following, visual reasoning, and scene-level generalization. In robotic manipulation, scaling VLA fine-tuning across multiple robot setups–especially when unifying representations across embodiments–has been shown to improve in-dataset performance and cross-embodiment generalization; in autonomous driving, however, VLAs remain largely trained on individual datasets and are rarely evaluated for zero-shot transfer to unseen datasets and camera rigs; furthermore naively adding more datasets to the training data does not necessarily lead to better performance within seen embodiments. To address these problems, we study multi-dataset training for the driving task and BEV-Forcing, an auxiliary objective that transfers ground-plane object-layout information from a specialized Bird's-Eye-View model into the VLA backbone. By encouraging the model to represent object position through a shared BEV spatial interface, we show that an auxiliary task such as BEV-Forcing can improve both in-distribution and out-of-distribution performance when training on a small number of camera rigs. As the number of training embodiments increases, however, the benefits of the auxiliary task are reduced; we present this as evidence that new techniques in the literature may see their benefits diminish when simply scaling up training diversity, which motivates presenting results taking into account data scaling.
Read the full abstract

All authors Caio Azevedo, Stefano Sabatini, Sascha Hornauer, Fabien Moutarde

Mesoscale and Nanoscale Physics

Room-temperature local strain control of moiré excitons in MoS$_2$/WSe$_2$ heterobilayers

Moiré superlattices in heterobilayers of atomically thin transition metal dichalcogenides provide a versatile platform for exploring quantum many-body physics as they can trap excitons, leading to the formation of quantized moiré exciton states. However, such moiré excitons have been predominantly studied at cryogenic temperatures, which severely limits their practical applications. Here, we demonstrate room-temperature activation and control of moiré excitons in MoS$_2$/WSe$_2$ heterobilayers using local strain engineering. Applying mechanical strain with a modified atomic force microscopy tip, we observe a series of resonances attributed to interlayer exciton states confined in the moiré potential. Power-dependent photoluminescence measurements elucidate the population dynamics of the moiré exciton states, while controlled local strain enables continuous tuning of their emission wavelength to the center of the second telecom window. We provide a theoretical model that captures all experimentally observed features, including the unconventional spectral shape of the moiré exciton emission. Our findings establish local tip-induced strain as a powerful tool for the on-demand manipulation of moiré excitons, paving the way for room-temperature quantum excitonic devices.
Read the full abstract

All authors T. O. Oskolkova, Y. Koo, A. A. Shubnic, I. Yu. Chestnov, I. Choi, C. Cheng, D. Huang, T. Jiang, K.-D. Park, I. V. Iorsh, V. Kravtsov

Machine Learning

AGI Maze Prediction Datasets: A Compact Benchmark for Learning World Dynamics with Transformers

World modeling requires a predictive model to maintain and update an internal state adequate for reasoning about the consequences of actions. We introduce the AGI Maze Prediction Datasets and Benchmark, a lightweight controlled testbed for studying this capability in Transformers and other predictive models. Derived from procedurally generated, stateful grid worlds, the benchmark comprises per-step transition prediction, fixed-horizon state prediction, and sequential textual-observation prediction. Source-maze-disjoint training and validation splits, together with greedy exact-match evaluation, distinguish learning transferable action-conditioned dynamics from memorizing transitions in familiar layouts. We establish from-scratch byte-level Transformer baselines and compare them with two working-memory-augmented architectures. A generic auxiliary latent-memory Transformer can fit some training sets perfectly but does not consistently improve held-out performance. In contrast, a pseudo-video spatial-memory Transformer initializes a two-dimensional latent workspace from the input map and updates it from action history without receiving intermediate maps, positions, or state labels. Under the same data, objectives, and evaluation protocol, this model reaches perfect validation accuracy on selected fixed-horizon tasks where the byte and unstructured-memory baselines do not, and substantially improves sequential text-trace prediction. These results suggest that structured, task-aligned working memory can be more useful than additional latent capacity alone. More broadly, we argue that language grounding is mediated by persistent data structures and computations over them; the benchmark offers a compact setting for testing architectures that couple textual interfaces to learned structured state.
Read the full abstract

Mesoscale and Nanoscale Physics

In silico design of magnonic lasing in constricted waveguides

Analogue black-hole systems have been proposed in various physical platforms, including magnetic materials, offering rich physics and promising applications such as wave lasing. However, their practical realization and characterization remain largely unexplored. Here, we present an in silico study of magnonic black-hole phenomena in constricted ferromagnetic waveguides driven by spin-polarized currents. Using micromagnetic simulations with a two-dimensional Poisson solver to obtain realistic current-density profiles, we demonstrate the formation of a double-hole cavity bounded by analogue event horizons, enabling resonant spin-wave amplification. We characterize the resonances as a function of magnetic and geometric parameters and identify the corresponding spin-wave modes. We further show that gradual tapers enhance transmission by suppressing spin-wave reflections, while interfacial Dzyaloshinskii-Moriya interaction can mimic the current-induced Doppler shift, substantially reducing the critical current density required for applications and, consequently, Joule heating. Finally, we demonstrate a spin-wave laser in which thermally excited spin waves undergo selective amplification and coherent emission at well-defined resonance frequencies. These results provide design principles for magnonic analogues of gravitational systems and point toward their potential for advanced spintronic applications.
Read the full abstract

Instrumentation and Detectors

Toward High-Resolution Detection of Target Fragmentation: TEA-Sensitized NIT for Proton Therapy Applications

Nano Imaging Trackers (NIT) are fine-grained nuclear emulsions capable of tracking charged particles with sub-micrometric spatial resolution. The DAMON (Direct Measurement of Target Fragmentation) experiment recently employed NIT to detect target-fragmentation events relevant to proton therapy in direct kinematics. These measurements showed that the small crystal size, while providing high spatial resolution, limits the reconstruction efficiency for primary proton tracks and high-energy secondary protons. This work reports an enhancement of NIT sensitivity achieved through alternative chemical sensitization methods, namely gold-plus-sulfur (Au-S) and triethanolamine (TEA), together with the use of the GR-1 developer. The detector response was evaluated with gamma-ray, carbon-ion and proton exposures, including clinically relevant proton energies.
Read the full abstract

All authors V. Boccia, T. Asada, A. Alexandrov, G. Ambrosi, S. Argirò, M. Barbanera, N. Bartosik, G. Battistoni, A. Bigot, M. G. Bisogni, G. Butella, F. Cavanna, P. Cerello, E. Ciarrocchi, N. D'Ambrosio, G. De Lellis, A. Di Crescenzo, B. Di Ruzza, M. Dondi, M. Donetti, Y. Dong, M. Durante, R. Faccini, V. Ferrero, C. Finck, E. Fiorina, M. Francesconi, M. Franchini, G. Franciosini, L. Galli, A. Iuliano, K. Kanxheri, B. Kharpuse, M. Kimura, S. Kodaira, A. C. Kraan, A. Lauria, E. Lopez Torres, T. Maggipinto, M. Magi, A. Manna, M. Marafini, S. Masci, M. Massa, C. Massimi, I. Mattei, S. Mazzolani, A. Mengarelli, A. Mereghetti, R. Mirabelli, A. Moggi, M. C. Montesi, M. C. Morone, M. Morrocchi, S. Muraro, T. Naka, N. Pastrone, V. Patera, F. Pennazio, C. Pisanti, P. Placidi, M. Pullia, F. Quattrini, S. Rabaglia, L. Ramello, C. A. Reidel, R. Ridolfi, L. Rocchetti, L. Sabatini, L. Salvi, C. Sanelli, A. Sarti, O. Sato, S. Savazzi, A. Schiavi, C. Schuy, E. Scifoni, L. Servoli, G. Silvestre, M. Sitta, K. Someya, B. Spadavecchia, R. Spighi, E. Spiriti, L. Testa, V. Tioukov, S. Tomassini, F. Tommasino, M. Toppi, G. Traini, A. Trigilio, G. Ubaldi, S. Valentinetti, M. Vanstalle, M. Villa, U. Weber, R. Zarrella, A. Zoccoli, G. Galati

Artificial Intelligence

SALA: Semantic-Aware Logical Alignment for Complex Reasoning in In-Context Learning

Effective in-context learning (ICL) for complex reasoning relies on selecting the right demonstrations. Traditional retrieval methods based on surface similarity fail to capture the underlying problem-solving logic. Recent logic-based methods address this by matching predefined reasoning steps, but the rigid rules and exact-match criteria is improper to handle flexible or diverse reasoning processes. To address the problem, we propose SALA, a Semantic-Aware Logical Alignment framework. Instead of relying on a fixed inventory, SALA automatically learns task-specific reasoning operations. It then embeds these operations into a continuous semantic space and uses dynamic time warping (DTW) to align the reasoning sequences. This approach allows for soft, flexible matching of reasoning logic while remaining highly interpretable. Experiments across four reasoning benchmarks and three LLMs demonstrate that SALA outperforms existing demonstration selection methods. Further analysis confirms the roles of the operation induction and the logical semantic alignment.
Read the full abstract

All authors Zhao Ji, Wenqing Chen, Zhixuan Chu, Jianxing Yu, Jingping Liu, Shanhe Zhao, Zibin Zheng

Optics

Dispersion-engineered meta-coverslip for multimodal synthetic imaging

Multimodal optical imaging provides comprehensive sample characterization but has traditionally been hindered by complex and bulky instrumentation. Here, we propose a dispersion-engineered meta-coverslip to seamlessly integrate bright-field, differential, fluorescence, and holographic imaging modalities within a standard microscope architecture, requiring no hardware modification or realignment. The meta-coverslip utilizes a scalable subwavelength multilayer film to engineer spatio-temporal dispersion for a customized high-dimensional transfer function. As an example, we demonstrate flexible switching among the four imaging modalities by simply tuning the illumination wavelengths. Lastly, we demonstrate that the synthesis of multimodal images can provide spatially registered structural and molecular information for more comprehensive biological analysis. Our approach provides an accessible, scalable, and flexible solution for advanced imaging and is extensible to other multiplexed optical systems in sensing and computing.
Read the full abstract

All authors Yuchen Ma, Danlin Xu, Xinwei Wang, Guangwei Hu, Liangcai Cao

Combinatorics

The Maximum Number of Shortest Paths in Graphs

Benjamini and Tzalik obtained an upper bound on the number of shortest paths between two vertices at distance $t$ in a multigraph of maximum degree at most $\Delta$, and proposed a conjecture on the sharp bound. In this paper, we develop a probabilistic counting argument based on probability distributions induced by random walks from the two endpoints. This approach yields a sharp bound for multigraphs and confirms their conjecture. We further determine the exact maximum for simple graphs and thus answer another question of Benjamini and Tzalik. We also investigate the equality cases, describing the structure of the subgraph formed by shortest paths between $x$ and $y$ and giving tight examples.
Read the full abstract

Computer Vision

ORB-SVM : An Innovative Hybrid Framework for Efficient Brain Tumor Detection from MRI Scans

Brain cancer remains one of the most significant challenges in modern medicine, where the accuracy of early stage diagnosis is a decisive factor in patient survival and treatment efficacy. Although Magnetic Resonance Imaging (MRI) is the established gold standard for visualizing neurological structures, the interpretation of these high dimensional scans is often complicated by subjective variability among practitioners and the inherent noise present in complex medical images. While contemporary approaches frequently rely on high parameter deep learning architectures, such models often involve significant computational costs and require extensive data for effective training. This study introduces a hybrid framework that utilizes the Oriented FAST and Rotated BRIEF (ORB) algorithm for precise feature extraction and a Support Vector Machine (SVM) for classification [1], [2]. The proposed approach achieves a sub- stantial data reduction of approximately 99.5%, which effectively minimizes the influence of non informative background data while preserving critical diagnostic patterns essential for tumor identification. By balancing feature sparsity with a robust kernel based classifier, this methodology addresses the limitations of over parameterized systems while maintaining high diagnostic integrity. Experimental evaluations conducted on the Br35H dataset demonstrate that the framework attains a classification accuracy of 97.5%. The findings suggest that the integration of localized feature representation and optimized classification provides a reliable and resource efficient alternative for medical image analysis, offering a structured solution that maintains per- formance without the need for extensive computational overhead.
Read the full abstract

Optimization and Control

A simple derivation of the Kalman filter

In this lecture note, we present a concise and self-contained derivation of the discrete-time Kalman filter equations that requires only a basic understanding of least squares estimation. The treatment is designed to minimize mathematical overhead while preserving both rigor and generality.
Read the full abstract

Probability

Metastability of interacting stochastic systems with countably many metastable states: beyond positive recurrence

Metastability is typically formulated for systems with finitely many metastable states, most often in positively recurrent settings. In this article, we extend the resolvent framework for metastability to Markov processes with countably many metastable states, thereby encompassing null-recurrent and transient dynamics. For a family of processes on locally compact Polish spaces, we prove, under a mild boundary regularity assumption, that the asymptotic flatness of microscopic resolvent solutions, supplemented by two compactness conditions, is equivalent to convergence in law of the projected trace processes to a limiting Markov chain and to the negligibility of the time spent outside the metastable sets. We apply this framework to two non-compact stochastic systems. First, we study a condensing inclusion process on a countably infinite, uniformly locally finite graph, in a setting where the process may be null recurrent or transient. We prove that the condensate location converges to a weighted random walk on the underlying graph, while the time spent away from the fully condensed configurations is negligible. Second, we study a small-noise one-dimensional Langevin diffusion with countably many stable equilibria, without assuming ergodicity. In the Eyring-Kramers time scale associated with the minimal energy barrier, we establish local equilibration inside each well, convergence of the well-index process to an explicit nearest-neighbor Markov chain on $\mathbb Z$, and negligibility of inter-well excursions. Together, these results broaden the scope of resolvent-based metastability theory beyond finite metastable state spaces and positive recurrence.
Read the full abstract

High Energy Physics (Theory)

QFT on flat, compact, orientable manifolds

Motivated by the recent observation that rotating space-times at finite temperature can be described by particular flat, compact, orientable manifolds, which are different from $\mathbb T^4$, we catalogue all such 4-manifolds and their properties. There are in total 26 flat, compact, orientable 4-manifolds, 23 of them spin-manifolds, so fermions and a Dirac operator can also be defined on these. Seven of them are appropriate for finite volume periodic boxes undergoing rotation at finite temperature.
Read the full abstract

High Energy Physics (Phenomenology)

Distribution amplitudes of vector and axial-vector mesons in a nonperturbatively improved symmetry-preserving framework

Using continuum Schwinger-function methods with a nonperturbatively improved, symmetry-preserving kernel, we deliver predictions for the leading-twist light-front distribution amplitudes (DAs) of the $\rho$, $K^\ast$, $a_1(1260)$, $b_1(1235)$, and the unmixed strange partners of the $1^{++}$ and $1^{+-}$ axial-vector (AV) channels, reconstructed from Mellin moments of the associated Bethe–Salpeter wave functions. For vector mesons, polarisation barely affects longitudinal momentum sharing: the longitudinal and transverse DAs are nearly degenerate, and both narrower than the asymptotic distribution in the second-moment sense. The AV sector is different in kind. Charge conjugation compels one projection – interchanged between the $1^{++}$ and $1^{+-}$ channels – to vanish at $x=1/2$ and change sign; breaking $SU_F(3)$ symmetry removes this protection, whereupon the zeroth moments become nonzero and the nodes shift from the midpoint. Under a common weighted normalisation, the $1^{+-}$ zeroth moment is $1.63$ times that of the $1^{++}$ channel, and the profile distortion follows the same pattern. A coupling forbidden by charge conjugation in the symmetric limit, $f_{b_1}=0$, becomes $f_{K_1^{+-}}=0.019\,$GeV in the strange channel: an independent measure of the same symmetry breaking, obtained from a current matrix element rather than from the DA reconstruction. What distinguishes the two sectors is thus a symmetry-enforced zero, not the size of the flavour asymmetry.
Read the full abstract

All authors Zhen-Ni Xu, Zhao-Qian Yao, Peng Cheng, Khépani Raya

Databases

Poisoning Attacks on the PGM-index

The PGM-index (Ferragina and Vinciguerra, VLDB'20) is one of the most practical learned indexes, owing to its theoretical elegance and consistently strong empirical performance. It is built on optimal piecewise linear approximations (PLAs) that minimize the number of segments. In this paper, we ask how sensitive this optimal PLA itself is to poisoning attacks. We propose PGM-attack, an efficient poisoning attack that sequentially inserts adversarial keys to inflate the resulting number of segments, and we develop a method for deriving theoretical upper bounds on the number of segments attainable under arbitrary insertions. Our experiments show that poisoning only 10% of the keys allows PGM-attack to increase the segment count by up to 120x. On every evaluated instance, our instance-dependent upper bound is at most 1.92x the segment count attained by PGM-attack, certifying that PGM-attack achieves at least 52% of the optimum. This increase in the number of segments enlarges the PGM-index by up to 120x. Moreover, the attack also transfers to other learned indexes, substantially inflating the index size of PLA-based ones in particular. Our results reveal that, despite the optimality of its PLAs, the PGM-index has an intrinsic vulnerability rooted in its optimization objective, motivating robustness-aware objective design for future learned indexes. Our code is publicly available at https://github.com/atsukisato/pgm-attack.
Read the full abstract

Number Theory

On the counting function of square-full numbers

We study the distribution of square-full numbers under the assumptions of Riemann hypothesis and that the zeros of $\zeta(s)$ are simple. Thus the error term is improved to a considerable extent with an extra main term.

Plasma Physics

Chirped-Pulse Forward Raman Amplification in Nonuniform Plasmas

Light amplification via Raman scattering in plasma has been severely constrained by stringent phase matching conditions and the need for plasma uniformity. To overcome these limitations, we propose a forward Raman amplification scheme that employs a positively chirped seed pulse co-propagating with a pump pulse in a nonuniform plasma with an upramp density profile. We demonstrate that the phase detuning induced by plasma nonuniformity can be dynamically compensated, enabling broadband amplification across the entire spectral bandwidth of the seed pulse. Concurrently, the chirped pulse duration undergoes continuous compression as a result of the spatially varying dispersion of the plasma. Our theoretical model, incorporating the detuning term and supported by particle-in-cell simulations, elucidates the compensation mechanism. It is shown that a chirped seed pulse with an initial bandwidth ~10% can be directly amplified by a factor of 1e7 to an intensity exceeding 1e17W/cm2 within a picosecond timescale in a steep density ramp. This scheme establishes a new foundation for advancing plasma-based light amplification toward practical applications.
Read the full abstract

All authors Zhi-Yu Lei, Zheng-Ming Sheng, Su-Ming Weng, Min Chen, Jie Zhang

Analysis of PDEs

A Payne-Weinberger inequality for quantum dot Dirac operators

In this work we prove a Payne-Weinberger type inequality for quantum dot Dirac operators defined on bounded and simply connected planar domains. This is a sharp upper bound for their first positive eigenvalue depending only on the isoperimetric deficit of the domain. To this end, we use a recently studied connection with the so-called $\overline\partial$-Robin Laplacian and we establish an analogous inequality for its first eigenvalue, relying on the corresponding inequality for the Robin Laplacian.
Read the full abstract

Information Retrieval

Adaptive Test-Time Inference for Text2Cypher with Trace Budgeting and Selective Refinement

Large language models have enabled natural language interfaces for structured databases, but generated queries may still contain syntactic errors, violate database schemas, or fail during execution. Test-time inference strategies improve generation reliability without additional training, but existing approaches often use fixed inference budgets and uniform refinement strategies, leading to unnecessary computation across questions with different complexity levels. In this work, we investigate adaptive test-time inference for Text2Cypher and introduce two strategies: adaptive trace budgeting, which dynamically adjusts the candidate generation budget based on question difficulty, and selective execution-guided refinement, which applies correction only when additional inference is expected to be beneficial. Experiments on Gemma-2-9B and Qwen-2.5-7B show that adaptive trace budgeting reduces the average generation budget by 30.7% and wall-clock inference time by 21-25% while maintaining comparable generation quality. Selective refinement preserves nearly all execution success gains of full refinement, reducing execution success by only 0.2-0.5% while avoiding unnecessary refinement for simpler questions. Experiments show that a single correction model (Gemma-4) effectively refines outputs from a different model family, suggesting refinement transfers across families.
Read the full abstract

Emerging Technologies

Quantum Workload Privacy Beyond Data Confidentiality

Remote quantum computing exposes a confidentiality gap. Standard privacy mechanisms protect quantum states and outputs, but not the scientific structure of a workload. This work reveals that hardware-aware compilation leaves observable signatures, such as routing overhead, circuit depth, and gate composition, that correlate with hidden modelling choices like partial differential equation boundary conditions, discretisation scale, and molecular geometry. The leakage arises from the mismatch between logical topology and fixed hardware connectivity, forcing problem-dependent SWAP insertion. We formalise this threat as Scientific-Intent Indistinguishability and prove that passive security is asymptotically unachievable under routing-optimal compilation. Experiments on a 156-qubit IBM Heron processor achieve near-perfect classification of boundary regimes and molecular geometries, with leakage generalising across solver families via routing-scaling exponents. Conventional gate-padding fails as a defence, causing fidelity drops without reducing adversarial advantage. Our results show that protecting quantum data alone is insufficient; execution-level confidentiality must become a first-class design requirement.
Read the full abstract

Machine Learning

What Is Worth Representing? Representational Empowerment for Continual Model Construction

The first problem of modeling the world is not just estimating the right parameters or causal structure, but deciding what should be represented at all. We frame this problem as continual model construction: an agent maintains an environment-specific model M of an inaccessible world W and curates a persistent library L of reusable representational elements across environments. We propose Representational Empowerment (RepEmp) to score candidate elements by how much they expand the agent's future capacity to model and plan, complementing the classic definition of empowerment, but redefined as control over internal representations instead of external states. We realize the framework as a hierarchical Curator-Actor architecture and test it across three experiments. In a closed-vocabulary causal-learning task, human participants construct causal models at varying abstraction granularities to maximize goal reachability rather than fidelity to the world, a signature better predicted by RepEmp than by information-gain alternatives. Matched simulations reveal that RepEmp-guided construction contributes more than exploration to sufficient structure recovery and cross-task transfer. Finally, in an open-vocabulary planning domain, an LLM-augmented Curator builds more compact symbolic libraries, which also generalize better than baselines. Ablating RepEmp eliminates these benefits. Together, these results identify RepEmp as a key principle for continual model construction: deciding what to build, retain, and reuse under bounded resources.
Read the full abstract

All authors Fei Dai, Hanqi Zhou, Alison Gopnik, Charley Wu

Materials Science

Ferroelastic domain switching by ultrafast photoinduced strain

The crystal lattice underpins the fundamental properties of condensed matter, with ferroic order emerging sensitively from atomic coordination. Manipulating the lattic therefore promises a direct way to switch ferroics between their states. Infrared excitation provides an efficient pathway to drive lattice motion and generate transient crystal deformations and strains. Ferroelastics constitute a uniquely direct platform for such control, their order parameter being strain itself. However, whether ultrafast laser-induced lattice distortions can switch ferroelastic order remains unknown. Here we use ultrafast pump-probe microscopy to demonstrate that a single high-amplitude infrared pulse induces strain within the first nanosecond, followed by ferroelastic domain switching several nanoseconds later. The spatial distribution follows that expected from the photoinduced strain field, while their temporal evolution is closely coupled to the strain dynamics. Our observations identify transient strain as the driving field for ferroelastic switching and suggest a general lattice-mediated pathway for controlling ferroic order using infrared light.
Read the full abstract

All authors Daniel G. Lourens, Maarten Kwaaitaal, Carl S. Davies, Andrei Kirilyuk

Probability

Analysis of Triggered Packet Streams: A Matrix-Analytic Method for Exponential Triggering Delays

In many communication networks, the transmission of a packet may automatically trigger the transmission of a subsequent packet from the same source after a (possibly random) delay, without requiring acknowledgment or feedback. Such behavior arises in multi-stage status updating, proactive protocols, and other applications where users generate causally dependent packet streams. In this paper, in order to analyze these systems, we introduce the $\mathrm{M^T/G/1}$ queue. In this model, primary customers arrive according to a Poisson process, and each primary customer triggers a secondary customer to join the queue after an independent delay. This arrival mechanism falls outside the scope of classical queueing models with renewal arrival processes. When the triggering delays follow an exponential distribution, we exploit the memoryless property to set up a tractable Markov description. By truncating the number of pending secondary customers, we derive a finite system of linear algebraic equations in the Laplace–Stieltjes transform domain and solve them using matrix-analytic methods. Based on the resulting workload distribution, we compute class-specific performance metrics using PASTA for primary customers and Palm conditioning for secondary customers. Finally, we validate the accuracy of this truncation through numerical experiments.
Read the full abstract

Robotics

An Open Panoramic Aerial Robot: Airframe-Integrated Multi-Fisheye Sensing, Onboard ERP Formation, and Field Evaluation

We present an open panoramic aerial robot with four synchronized fisheye cameras integrated into a carbon-fiber airframe and an onboard NVIDIA Jetson Orin NX. The robot outputs calibrated raw views and an equirectangular panorama (ERP; 1280x640 in all experiments). The ERP pipeline uses overlap-specific projection radii, gated local alignment, seam control, and multi-rate state updates, and it runs onboard on the live four-camera stream during flight. The field dataset contains 18 sequences and more than 50,000 synchronized groups from seven sites. On a 60-frame far-field sample, the method reduces the median per-frame AKAZE P90 misalignment by 40.7% compared with Fixed Radius, and with fixed parameters it gives the lowest geometric errors among the tested controls at two held-out sites. Controlled replay on the same NVIDIA Jetson Orin NX measures final-ERP continuity, timing, and module-input power at a 20 Hz input rate. Compared with external stitching software given the same calibrated projection, the onboard pipeline gives final-ERP line continuity no lower than any tested method, while every external configuration measured on the module needs 5.3 to 147 times the input period and 3.8 to 106 times the energy per output. Frozen detection and place-recognition models are used to evaluate the exported images. Code, calibration, reference hardware, and data-access documentation are available in an anonymized repository at https://anonymous.4open.science/r/Open-Pano-Field-CE1F/README.md.
Read the full abstract

All authors Dun Dai, Ze Lu, Cheng He, Yaowen Wang, Quan Quan