Peer Revenue Returns subject to peer review

Papers

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

All of arXiv

Tuesday 1 September

Solar and Stellar Astrophysics

Revisiting candidates for non-pulsating stars located in the Cepheid instability strip in the Large Magellanic Cloud

We analyzed photometric and spectroscopic data for 11 candidates for non-pulsating stars located in the Cepheid instability strip (IS) of the Large Magellanic Cloud (LMC) in order to investigate the reasons for the lack of pulsations. We used available temperature calibrations based on photometric colors to estimate the effective temperatures of the candidates, which served as initial parameters for the spectroscopic analysis. We also applied surface brightness-color relations calibrated for Cepheid variables, giants, and supergiants to estimate stellar radii. The spectral analysis was performed using the spectral synthesis method to determine the atmospheric parameters; namely, the effective temperature, metallicity, surface gravity, microturbulent velocity, projected rotational velocity, and chemical abundances for up to ~30 elements. For most objects, only a single spectrum was available. However, no significant variations in radial velocities were detected among the stars with repeat observations; therefore, all stars were treated as single object in the analysis. Two stars exhibit broad spectral lines, which may indicate high rotational velocities or possible binarity; however, additional spectra are required to confirm this interpretation. One of these objects also shows asymmetric line profiles, which might be related to the presence of non-radial pulsation modes causing line-shape variations. A common feature among all analyzed candidates is an enhancement of barium-peak s-process elements compared to solar values. This may indicate past mass transfer from a companion during its post-asymptotic giant branch phase. This study provides an insight into the physical parameters of candidates for non-pulsating stars residing within the Cepheid IS in the LMC, although the lack of pulsations remains a mystery and challenges pulsation theory.
Read the full abstract

All authors W. Narloch, C. Galan, G. Pietrzynski, B. Pilecki, W. Gieren, R. Smolec, H. Netzel, P. Wielgorski

Optimization and Control

Sum-of-Squares Certificates for Copositive Matrices via Recursive Identities: The de Klerk-Pasechnik Conjecture and Hoffman-Pereira Matrices

We establish the conjecture by de Klerk and Pasechnik (2002), claiming that the semidefinite bounds $\vartheta^{(r)}(G)(r\geq 0)$ for the stability number $\alpha(G)$ are exact at $r=\alpha(G)-1$, by exhibiting an explicit sum-of-squares certificate. This certificate allows us to recover a known characterization of the minimizers of the Motzkin-Straus formulation for $1/\alpha(G)$. Additionally, we give sum-of-squares copositivity certificates for the matrices satisfying the Hoffman-Pereira sign condition, a crucial condition for characterizing copositive matrices with $\{-1,0,1\}$ entries.
Read the full abstract

Numerical Analysis

OptiXDE: A fast optical-inspired solver for differential equations

OptiXDE is a matrix-free spectral operator framework for differential equations on uniform grids and embedded domains. Inspired by angular-spectrum propagation in Fourier optics, it maps transform-diagonal spatial operators to analytical modal multipliers and composes them with physical-space operators for nonlinearities, geometry and boundary enforcement. A common transform–operator–inverse-transform backbone is demonstrated across transient diffusion, periodic and embedded-domain Poisson problems, the cubic nonlinear Schr"odinger equation, viscous Burgers dynamics, the two-dimensional Allen–Cahn equation and incompressible flows from the Taylor–Green vortex to embedded-cylinder vortex shedding. Transform-compatible linear problems are recovered near the floating-point limit, whereas errors on the singular L-shaped domain remain localized near the re-entrant corner and regularized interface. Nonlinear benchmarks recover second-order temporal convergence and the expected conservative or dissipative behavior, while incompressibility remains near round-off level during long-time vortex shedding. The matrix-free updates require \(\mathcal{O}(N\log N)\) work and \(\mathcal{O}(N)\) memory. Device-resident transform workloads reach \(94.9\times\) GPU acceleration, and the complete embedded-cylinder solver achieves a \(42.1\times\) CPU–GPU speedup under matched numerical settings. These results establish OptiXDE as a deterministic and extensible operator-centric framework for structured and embedded-domain differential equations.
Read the full abstract

Quantum Physics

Uniform Hiding and Two Routes to Relative Accuracy in Gaussian Boson Sampling

Gaussian boson sampling requires control of how closely finite optical matrices follow Gaussian reference laws. We prove an explicit total variation bound of order $N^2/M$ between a rescaled Haar transpose Gram block and its Gaussian transpose Gram counterpart, where $N$ is the detected photon count and $M$ is the number of optical modes. The bound establishes quantitative product hiding uniformly over every number of squeezed inputs. The proof combines Stiefel recursion, centered circular orthogonal ensemble scores, and a rectangular entropy estimate. Applications combine hiding with local hafnian bounds to obtain relative probability guarantees and connect them to sampler error through exact photon-sector normalization.
Read the full abstract

Probability

Global universality of the expected number of zeros of non-analytic random signals

We study the asymptotics as $n$ goes to infinity of $\mathbb E\left[\mathcal{N}(S_n,[0,2\pi])\right]$, the expected number of zeros in $[0, 2\pi]$ of a random periodic signal $S_n$ of the form \[ S_n(t)=\sum_{k=1}^{n}a_k f(kt), \] where $f$ is a non-analytic $2\pi-$periodic function and the coefficients $(a_k)$ are i.i.d. random variables, centered with unit variance. We show in particular that if $a_1$ admits a finite third moment and if the function $f$ is piecewise polynomials and of class $\mathcal C^{7}$, then we have the following universal asymptotics, independent of the particular law of the coefficients $(a_k)$ \[ \lim_{n \to +\infty}\frac{\Esp\left[\mathcal{N}(S_n,[0,2\pi])\right]}{n}= \frac{2}{\sqrt{3}}\sqrt{\frac{\|f'\|_{L^2([0,2\pi])}}{\|f\|_{L^2([0,2\pi])}}}. \] This result thus extends in expectation and at the scale of the whole period $[0,2\pi]$ the local universality property established {in [Angst-Poly, IMRN, 2019]}, in distribution and in shrinking intervals of size $1/n$. Moreover, it generalizes to a non-analytic context the global universality results obtained in the more classical frameworks of random trigonometric polynomials or random analytic functions. Our approach combines a new almost sure Central Limit Theorem à la Salem–Zygmund for the function $S_n$ when evaluated at a uniform random point in $[0, 2\pi]$, and as well as suitable uniform integrability and anti-concentration estimates.
Read the full abstract

Artificial Intelligence

Figures as Programs: Recursive Generation of Editable Scientific Figures

Scientific methodology figures are essential for communicating complex methods clearly, yet creating them remains labor-intensive and typically requires multiple rounds of refinement. Recent image-generation models can synthesize visually appealing raster figures, but producing a human-satisfactory result in a single generation step remains difficult. Moreover, precise edits to raster figures are challenging for both humans and models. We formulate scientific figure generation as recursive SVG program construction and propose FigTree, a multi-agent system that automatically transforms a scientific paper into a structured vector figure. FigTree grounds figure content in the source paper, decomposes a figure into a hierarchy of local regions, generates each region as a short SVG program, and assembles the resulting fragments. A render-critic refinement loop jointly inspects the rendered figure and its underlying program, enabling visual defects to be traced to specific statements and accurately repaired. We conduct extensive evaluations of FigTree on figure quality and editability, showing that FigTree produces high-quality figures, while also enabling more effective editing than existing raster-based methods.
Read the full abstract

All authors Yepeng Liu, Dasen Dai, Chengzhi Liu, Yiren Song, Hai Ci, Yu Zhang, Qi Zhang, Mike Zheng Shou, Xin Eric Wang, Yuheng Bu

High Energy Physics (Phenomenology)

Update of the Standard-Model prediction for $\bar B \to X_s \gamma$

We report on two recent calculations on the inclusive radiative decay $\bar B\to X_s \gamma$, notably multi-parton contributions at NLO and the $Q_{1,2}-Q_7$ interference at NNLO for the physical value of the charm-quark mass. The former calculation formally completes $\bar{B} \rightarrow X_s \gamma$ at NLO in QCD at leading power, the latter removes a long-standing $\pm 3\%$ uncertainty arising from interpolation in $m_c$. The updated Standard-Model prediction for the CP- and isospin-averaged branching ratio reads ${\mathcal B}_{s \gamma}^{\rm SM} = (3.54 \pm 0.14)\times 10^{-4}$ for photon energies $E_\gamma > 1.6\,{\rm GeV}$ in the $B$-meson rest frame. This value is in good agreement with the current experimental average ${\mathcal B}_{s \gamma}^{\rm exp} = (3.49 \pm 0.19)\times 10^{-4}$.
Read the full abstract

Computer Vision

SinkPruner: Sink-Free Visual Token Pruning for Multimodal Large Language Models

Despite their strong multimodal understanding ability, multimodal large language models (MLLMs) incur substantial computational overhead when processing long visual token sequences. To reduce inference costs, recent studies have explored visual token pruning through vision-centric or text-guided strategies. However, these methods often overlook high-norm outlier tokens, i.e., tokens with abnormally large feature norms, leading to suboptimal pruning decisions. In this work, we show that such high-norm outlier tokens are highly redundant in both feature and spatial dimensions, yet are often mistakenly preserved as informative cues by existing methods. Motivated by this observation, we propose SinkPruner, a training-free visual token pruning framework for efficient MLLM inference. SinkPruner follows a coarse-to-fine design with two key modules: a visual sanitizer that filters high-norm redundancies and alleviates attention sink and attention dispersion, and a text-guided pruner that further retains tokens semantically aligned with the text query. Extensive experiments on twelve image-language and four video-language benchmarks demonstrate the effectiveness, efficiency, and generalizability of our framework. Notably, SinkPruner preserves 96.5% (91.8%) of the original performance of LLaVA-1.5 (Qwen2.5-VL) under an 89% token reduction. Experiments further indicate that our visual sanitizer exhibits promising transferability in enhancing the performance of existing pruning methods. Our code is available at https://github.com/LaVi-Lab/SinkPruner.
Read the full abstract

All authors Shiyu Li, Zi-Yuan Hu, Shijia Huang, Yanyang Li, Yiwu Zhong, Liwei Wang

Cosmology and Nongalactic Astrophysics

A Class of Exact Single-Field Inflationary Solutions beyond Slow Roll

We construct exact solutions for single-field inflaton dynamics without invoking the slow-roll approximation. A suitable change of variables reduces the background equation to an Abel equation of the first kind. Although a generic Abel equation is not analytically solvable, we identify a class of inflaton potentials for which the transformed equation admits exact solutions. The resulting framework contains constant-roll inflation as a special case and also accommodates solutions with a constant second Hubble-flow parameter. We analyze the linear local attractor behavior and superhorizon evolution of these rolling backgrounds. Using the public joint CMB likelihood contours in the $(n_s,r)$ plane, we identify compatible parameter regions and show that one rolling branch can also yield $50\leq N_*<60$. Direct numerical evolution of the scalar and tensor modes at representative points validates the local-index predictions to better than $7\times10^{-4}$ in $n_s$ and $2\times10^{-6}$ in $r$. The exact family extends beyond slow roll, although the observationally selected regions displayed here lie close to the slow-roll regime.
Read the full abstract

All authors Jia-Wei Zhang, Bai-Cian Ke, Yao Yu, Dong-Ze He, Shou-Jia Wang

Materials Science

Numerical study of the grain-growth-induced non-parabolic kinetics of a solid-state reaction

Non-parabolic growth of reaction layers is frequently observed in solid-state diffusion couples, but the relationship between microstructural evolution and the resulting kinetics remains incompletely understood. This effect is commonly attributed to grain growth, which can progressively reduce the contribution of fast grain-boundary diffusion, causing the effective diffusivity of a polycrystalline product layer to evolve during reaction. This work develops a moving-boundary diffusion model to describe product-layer growth while accounting for the local grain-growth history of the continuously formed product, as well as both bulk and grain-boundary diffusion. Numerical simulations show that the reaction can pass through three distinct kinetic regimes: an initial approximately parabolic regime dominated by grain-boundary diffusion, a transient sub-parabolic regime associated with strong spatial variation of the effective diffusivity, and a subsequent approximately parabolic regime dominated by bulk diffusion. The magnitude and duration of the sub-parabolic regime depend on the relative grain-boundary and bulk diffusivities and on the kinetics of grain growth. The instantaneous growth exponent therefore evolves continuously and does not represent a unique kinetic constant. Nevertheless, fitting simulated layer-thickness data over finite experimental time intervals produces well-defined apparent exponents, demonstrating how a transient process can appear to obey a single power law.
Read the full abstract

Optimization and Control

Further analysis and extension of the higher-order Newton method of Ahmadi, Chaudhry, and Zhang

We extend a $d$th-order Newton method for unconstrained optimization by Ahmadi, Chaudhry, and Zhang [Advances in Mathematics, 452:109808] to optimization with SOS-convex polynomial constraints. Consider the problem of minimizing a smooth function $f:\mathbb{R}^{n}\to\mathbb{R}$ subject to SOS-convex polynomial constraints. Given an iterate $x\in\mathbb{R}^n$, Ahmadi et al. define the next iterate $x^{+}$ as the minimizer of the $d$th-order Taylor expansion of $f$ at $x$ with a regularization term of degree $d^{\prime}$, where $d^{\prime}$ is the smallest even number greater than $d$, chosen such that this polynomial is SOS-convex, subject to the constraints. Constructing this polynomial and minimizing it subject to the constraints can both be reduced in time polynomial in $n$ to a semidefinite program (SDP). We prove that, if $f$ is strongly convex and the tensor of the $d$th-order partial derivatives of $f$ is Lipschitz continuous, then our method converges locally to the optimal solution $x^{\ast}$ with order $d$. We further prove that, under certain constraint qualifications, the set of active constraints at $x^{\ast}$ is identified locally in a single iteration. Next, we study the worst-case performance of the third-order Newton method in the unconstrained setting for two classes of univariate $f$ using performance estimation. Finally, we extend a globally convergent modification of the $d$th-order Newton method to the setting of SOS-convex polynomial constraints.
Read the full abstract

High Energy Physics (Phenomenology)

NNLL fit of the transverse-momentum-dependent distribution of unpolarised gluons to LHCb data on $J/\psi$-pair production

We perform the first fit of the unpolarised gluon transverse-momentum-dependent distribution $f_1^g$ up to next-to-next-to-leading-logarithmic accuracy to the double-differential distribution in the invariant mass and transverse momentum of the latest $J/\psi$-pair production LHCb data. We review previous attempts to describe the older $J/\psi$-pair LHCb data using $f_1^g$, and discuss two novel methodologies which allow us to account in the fit for both the perturbative and PDF uncertainties to construct $f_1^g$ in the perturbative region. It turns out that the consideration of these uncertainties is crucial for a meaningful fit to such low-scale gluon-sensitive data.
Read the full abstract

All authors D. Boer, J. Bor, A. C. Serri, M. G. Echevarria, J. P. Lansberg, S. F. Romera, P. Taels

Computation and Language

Right Frame, Wrong Rule: Cultural Cues Expose the Financial Knowledge Gap They Were Meant to Close

When a question has valid answers under different normative frameworks, a language model must decide which framework to use and whether it can answer correctly within it. We call this setting normative pluralism and study it in Islamic finance using a four-choice taxonomy that separates framework selection from within-framework correctness. This separation reveals the stereotype trap: a cultural cue steers a model toward one framework, but the model selects an incorrect answer within that framework. Across twelve models, two languages, and fifty demographic signals, cultural cues change framework selection and reveal substantial differences in accuracy, especially among non-frontier models. Under the strongest signal, large open-weight models select the Islamic framework 97% of the time. A two-choice evaluation would report near-perfect alignment, although 57–66% of those selections are incorrect. These findings motivate, but do not directly test, the competence-conditioned routing hypothesis: models may favor frameworks where they are more accurate, while cultural cues may expose framework-specific competence gaps.
Read the full abstract

All authors Rania Elbadry, Ahmed Heakl, Saeed Almheiri, Fan Zhang, Muhra AlMahri, Xueqing Peng, Mohsinul Kabir, Shuyao Wang, Yi Han, Saadeldine Eletter, Duzhen Zhang, Preslav Nakov, Yuxia Wang, Fajri Koto, Zhuohan Xie

Computation and Language

Inspicio: Open-Vocabulary, LLM-Based Sense Retrieval for Historical Languages

Word Sense Disambiguation has advanced rapidly for English and a handful of well-resourced modern languages, but it continues to assume the existence of a sense inventory and a word-to-sense mapping in the source language (Navigli, 2026). These assumptions break down for most historical and low-resource languages, whose dedicated WordNets are either incomplete or still under construction. We present Inspicio, an open-vocabulary retrieval pipeline that links tokens in context to synsets of the Open English WordNet (McCrae et al., 2020) without requiring any source-language inventory or mapping. For each occurrence, an instruction-tuned LLM produces two English translations of the surrounding sentence, a small set of candidate dictionary-style definitions, and a few candidate English lemmas. These outputs drive a hybrid retrieval step that combines dense definition-synset similarity, sparse lemma matching, and Maximal Marginal Relevance re-ranking. We evaluate the pipeline across a 6x6 grid of LLMs and sentence-embedding models on a new bilingual set of manually annotated Latin and Ancient Greek perception verbs, on a subset of PREMOVE dataset (Farina, 2025), and on a diachronic sample of Italian. The best configuration reaches 96% Recall@50 on the perception-verb test set, with each component contributing measurable gains, and remains competitive in the out-of-domain and cross-lingual settings.
Read the full abstract

Metric Geometry

The Truncated Octahedral Conjecture

Among three-dimensional parallelohedra of fixed volume, the Archimedean truncated octahedron uniquely minimizes surface area, affirming a conjecture of K. Bezdek from 2006. The final scalar inequality is verified by an exact computer-assisted certificate.

Computer Vision

Does This Moment Justify the Recommendation? Counterfactual Behavior-Grounded Evidence Retrieval for Personalized Video Recommendation

Personalized video recommendation predicts user preference at the video level, while temporal video grounding localizes query-relevant moments. However, strong localization does not establish whether the retrieved moment constitutes valid evidence for recommending the video to a particular user. We study counterfactual behavior-grounded evidence retrieval, which separates where personalized evidence occurs from whether such evidence exists and evaluates whether model predictions respond consistently when that evidence is replaced. We introduce CBGER-10K, containing 5,000 controlled factual–counterfactual pairs for 3,026 users, where each pair replaces only the focal behavior-supported segment while preserving the user, temporal position, and hard distractors. We further propose CBGER, a compact framework that decouples segment-level localization from video-level evidence estimation and learns both through structured counterfactual supervision. CBGER achieves $0.4432$ MRR, $0.6977$ Pair Accuracy, and $0.6987$ Intervention Consistency across five adapted personalized-highlight and temporal-grounding baselines. Notably, compared with QD-DETR, its MRR improvement is not statistically significant, while Pair Accuracy improves by $11.03$ points. These results show that accurate temporal localization does not necessarily imply reliable personalized evidence existence, motivating explicit evaluation of Whether alongside Where.
Read the full abstract

Computer Vision

CQF-HMR: Continuous Quaternion Flows for Probabilistic 3D Human Mesh Recovery from a Single Image

Recovering 3D digital humans from a single 2D image is an ill-posed computer vision problem due to the loss of depth information. Probabilistic 3D human pose estimation compensates for this by estimating a set of 3D hypotheses from a prior distribution via generative models. However, most prior work focuses only on 3D keypoints, which often leads to implausible poses that are difficult to apply to downstream tasks, e.g. animation or digital humans. SMPL-based methods are more scalable thanks to the explicit body priors, but it requires more complex modeling of the generation process due to the non-additive nature of the joint rotations. In this work, we propose a novel approach for probabilistic 3D humans using quaternion-constrained continuous normalizing flows conditioned on 2D pose estimations. Our proposed quaternion flows show significant advantages over approaches using other rotation representations. Experiments demonstrate state-of-the-art results of our method on Human3.6M, particularly in ambiguous settings, and comparable pose estimation accuracy on challenging 3DPW and EMDB benchmarks.
Read the full abstract

All authors Cuong Le, Bao-Long Tran, Pavlo Melnyk, Tahereh Dehdarirad, Bastian Wandt, Mårten Wadenbäck

Computer Vision

EvoGS: Modeling Deformation Evolution for Dynamic Gaussian Splatting

Recent extensions of 3D Gaussian Splatting (3DGS) enable real-time novel view synthesis in dynamic scenes by learning time-conditioned Gaussian deformations. However, existing MLP-based methods typically estimate deformations independently at each timestamp, making them less robust to large or abrupt motions. To address this issue, we propose EvoGS, a 3DGS-based dynamic reconstruction framework that models Gaussian deformation as a temporal evolution process. EvoGS maintains persistent deformation states for each Gaussian, extrapolates future states from historical deformation states, and corrects the predictions with MLP-derived observations. The correction is adaptively weighted using a temporal residual memory and evolution statistics such as deformation velocity and trajectory deviation. To further improve reconstruction quality, EvoGS introduces deformation-aware densification. Clone and split operations are performed along corrected deformation directions, while an uncertainty-aware strategy suppresses densification for Gaussians with unstable deformation histories. Experiments show that EvoGS improves dynamic novel view synthesis quality and achieves competitive performance across benchmarks.
Read the full abstract

All authors Wei Dong, Shahram Shirani, Jun Chen, Han Zhou

Distributed Computing

AInfer-PD: Communication-Safe In-Place Prefill-Decode Multiplexing for Distributed MoE Rollouts

In agentic reinforcement learning (RL), asynchronous multi-turn trajectories repeatedly introduce continuation prefills (P) while other trajectories remain in decode (D). Efficient rollout completion therefore requires D to advance during long P executions. In-place multiplexing shares devices, model weights, and KV state, but phase-specific communication in distributed MoE introduces two constraints: cross-rank collective ordering and ownership of mutable communication state. Whole-prefill coordination can ensure safe ordering but delays D. DeepEP's normal and low-latency paths suit large P and small D batches, respectively, but safely overlapping them requires separating their mutable communication state. We present AInfer-PD, which extends in-place P/D multiplexing to distributed MoE rollouts. AInfer-PD establishes rank-consistent collective ordering at safe communication boundaries, allowing successive D iterations to advance without waiting for an entire prefill execution to be enqueued or completed. It gives P and D separate communication buffers and protocol state within one DeepEP runtime, allowing P to use the normal path while D retains the low-latency path. Together, these mechanisms preserve useful P/D overlap while sharing model weights and KV storage on the same devices. Across repeated single-node prefill-intensive workloads, AInfer-PD reduces fixed-workload rollout completion time by 7.1-22.5% relative to the same AInfer engine with P/D multiplexing disabled and by 24.8-32.9% relative to SGLang. On two nodes, the reductions are 18.0-35.3% and 18.3-31.8%, respectively. In a same-engine ablation, fine-grained boundaries reduce completion time by a further 8.6-19.8% over whole-prefill asynchronous enqueue.
Read the full abstract

All authors Guowei Wang, Chaokun Yang, Zhenxuan Pan, Yipeng Wei, Yuhong Guo, Minghua Zhu, Zechuan Zhang, Shuo Wan, Xiaowei Zhu

General Relativity and Quantum Cosmology

Dissipative Multi-Field Dynamics from Non-Hermitian Inflationary Potentials

In this work, we develop a perturbative framework for inflation driven by a complex inflaton with non-minimal gravitational coupling and a non-Hermitian potential. During the observable cosmic microwave background radiation era, the dynamics reduce to an effectively conservative two-field model, preserving the predictions of the $\alpha$-attractor class and satisfying Planck 2018 and BICEP/Keck constraints on $n_s$, $r$, and $f_{\mathrm{NL}}$. Near the end of inflation, trajectory bending activates the non-Hermitian sector, triggering geometric reheating. The resulting non-unitary evolution modifies the curvature spectrum and stochastic gravitational-wave background through a calculable damping factor determined by the complex mass eigenvalues. While cosmic microwave background-scale observables remain essentially unchanged, a distinctive suppression emerges in the high-frequency gravitational-wave spectrum ($f > 10^2$ Hz), potentially testable by future detectors such as the Einstein Telescope and the Big Bang Observer.
Read the full abstract

Atomic Physics

Impact of light shift inhomogeneities on the contrast of light pulse atom interferometers

We study the loss of the contrast in an atom interferometer when increasing the duration of Raman mirror pulses, and find the contrast decay rate to increase with the interferometer duration. We attribute this effect to the transverse spatial fluctuations of the intensity across the Raman beams, and to the dephasing induced by the associated light shifts inhomogeneities. Simulations based on the propagation of noisy synthetic Raman beams show that the contrast decay rate increases with the distance between the wavepackets at the mirror pulse, before reaching an asymptotic limit when intensity fluctuations between the two wavepackets become completely decorrelated. Finally, simulations based on the propagation of Raman beams having their measured intensity fluctuations predict contrast loss rates consistent with our measurements, confirming the detrimental role for the interferometer contrast played by intensity fluctuations across the interferometer laser beams.
Read the full abstract

All authors Maxime Pesche, Diego Lancheros Naranjo, Rayan Si-Ahmed, Gabriel Ducasse, Franck Pereira dos Santos, Sébastien Merlet

High Energy Astrophysical Phenomena

A Galactic microblazar as a potential accelerator of ultra-high-energy particles

Context. Persistent jets from X-ray binaries which are aligned very close to the line of sight could be considered to be Galactic equivalents of blazars, or 'microblazars'. They are also expected to power gamma-ray sources. Aims. We intend to assess a serious candidate apparently fulfilling many of the requirements to be considered a genuine member of this class: IRAS 18293-0941. Methods. An intense multi-wavelength observational and theoretical study has been carried out on our proposed candidate source. Results. With photometric and spectroscopic properties typical of a binary star, this system exhibits clear collimated and one-sided radio emission matching the effects of relativistic motion along a reduced ejection angle. Only fast variability is not observed possibly smoothed by a dense circumstellar envelope. A physical scenario is consistently modeled that also gives credibility to its likely connection with LHAASO J1831-1007u*, an ultra-high-energy source in its immediate vicinity. Conclusions. Our reported identification not only helps to fill a gap in Galactic taxonomy, but also potentially strengthens the role of the microblazar and microquasar families in our understanding of the most energetic Milky Way phenomena.
Read the full abstract

All authors Josep Martí, Pedro L. Luque-Escamilla, Benito Marcote, Leandro Abaroa, Arnau Aguasca-Cabot, Jorge A. Combi, Gustavo E. Romero, Josep M. Paredes, Federico García, Federico Fogantini, Enzo A. Saavedra, Daniel del Ser, Jakob van den Eijnden

High Energy Astrophysical Phenomena

The Missing MeV Window for Identifying Galactic PeVatrons: the need for a dedicated MeV Observatory

The identification of the sources responsible for accelerating Galactic cosmic-rays up to PeV energies remains a major open question in high-energy astrophysics. Recent observations by the Large High Altitude Air Shower Observatory (LHAASO) have revealed a growing population of Galactic PeVatron candidates. However, the detection of TeV-PeV gamma-rays alone does not uniquely identify the nature of the emitting particles. In many cases, both hadronic and leptonic scenarios can reproduce the observed very-high-energy emission, leaving the origin of the radiation unresolved. The most robust and least model-dependent electromagnetic tracer of hadronic acceleration is provided by neutral-pion decay, which produces the characteristic pion bump in the 10-100 MeV energy range. Observations in this energy band are therefore essential to establish whether candidate PeVatrons are accelerating protons and nuclei and, consequently, to assess their role in the origin of Galactic cosmic-rays. More broadly, the MeV domain addresses key astrophysical questions, but remains one of the least explored regions of the electromagnetic spectrum. As current and future VHE facilities expand the Galactic PeVatron census, the lack of MeV observations is becoming a major limitation for source identification. We discuss the astrophysical science drivers and observational requirements for a dedicated MeV gamma-ray capability optimized for pion-bump studies. We explore the concept of a compact, focused, and potentially fast-track mission capable of filling the impending MeV observational gap while more ambitious next-generation MeV facilities are developed and evaluated. Such a mission would provide the missing diagnostic needed to distinguish hadronic and leptonic emission, identify Galactic cosmic-ray accelerators, and fully exploit the scientific return of the emerging generation of TeV observatories.
Read the full abstract

All authors Martina Cardillo, Riccardo Campana, Yuri Evangelista, Fabio Muleri

Number Theory

Covering Projective Height Balls by Subspaces in Rigid Adelic Spaces

Let $E$ be an $n$-dimensional rigid adelic space over a number field $K$. We study the minimum number $g_E(R)$ of proper $K$-subspaces needed to cover the projective height ball of radius $R$, together with the maximum cardinality $h_E(R)$ of a subset in linear general position. We show that, once $R$ is sufficiently large compared with the last Roy–Thunder minimum of $E$, both quantities have order $\Psi_E(R)^{[K:\mathbb Q]}$, where $\Psi_E(R)$ is an explicit expression in the Roy–Thunder minima. The comparison constants are effectively computable and uniform in $E$. For the standard adelic space $K^n$, this gives order $R^{[K:\mathbb Q]n/(n-1)}$.
Read the full abstract

Sound

On the Human and Computer Alignment of Attribute-Based Music Matches

Recent advances in generative AI are raising ethical concerns regarding the originality of generated content and the potential replication of training data, with further implications for transparency, attribution, and intellectual property. In music, several computational approaches have been proposed to identify potential replication, using audio-based similarity metrics. Yet, their alignment with human judgments across distinct musical attributes remains underexplored. To address this gap, we conduct a perceptual experiment on music matches, defined as strongly similar musical excerpts. We focus on five musical attributes: melody, harmony, rhythm, voice, and timbre. We design a triplet-based forced-choice task comprising 300 cases, including plagiarism examples, cover songs, and AI-generated music. From this experiment, we introduce the MATCHA (Musical Attribute-based Triplet Comparison with Human Annotations) dataset: a collection of 1105 perceptual assessments of attribute-based music matches from 83 expert participants. Our findings reveal measurable agreement among participants in identifying matches across attributes. We further observe partial alignment between human judgments and computational similarity measures. Overall, this work underscores the importance of domain-specific and perceptually grounded evaluation frameworks for generative AI in creative practice.
Read the full abstract

All authors Roser Batlle-Roca, Woosung Choi, Joan Serrà, Fabio Morreale, Wei-Hsiang Liao, Xavier Serra, Emilia Gómez, Yuki Mitsufuji

Information Retrieval

TGR: Advancing Industrial Recommendation from Generative-Paradigm Ranking toward Unified Generation and Reasoning

Industrial recommender systems typically rely on cascaded retrieval, pre-ranking, ranking, and reranking stages, whose separately optimized models limit scaling, fragment decision making, and lack semantic knowledge and reasoning. We present TGR (Tencent Generative Recommendation), an industrial framework that advances recommendation toward the generative paradigm along three coupled directions. TGR-GenRank upgrades ranking through CCFormer, which combines unified feature tokenization, a scalable Transformer backbone, feature-field separated cross attention, subspace token mixing, and hierarchical sequence compression while retaining per-item multi-task outputs. TGR-GenRec explores end-to-end generation under two paradigms: BARGE bridges item-boundary loss and semantic drift in hierarchical semantic-ID generation through item context-aware attention, hierarchical path reranking, and orthogonal dual-path decoding; HiGR performs whole-slate generation with prefix-structured semantic IDs, coarse-to-fine decoding, and listwise multi-objective alignment. TGR-Reason injects offline-generated semantic-ID reason tokens into online decoding, providing reasoning without request-time rollout. TGR is deployed across Tencent production surfaces serving hundreds of millions of users. CCFormer delivers significant gains in five A/B-tested scenarios and is fully launched in two, including +3.57% CTR and +1.71% advertising revenue. BARGE improves Hit@5 by 10.2-16.9% and yields +0.60% CTR and +1.70% reading time after full rollout. HiGR improves offline slate quality by 15.9-21.3% with a 5x inference speedup and achieves up to +1.22% watch time and +1.73% video views. TGR-Reason raises cold-start new-user Hit@1 by 477.8% and delivers +1.75% effective consumption and +13.09% new-user exposure-to-conversion online.
Read the full abstract

All authors TGR Team: Lei Cheng, Haonan Hu, Beibei Kong, Yudong Li, Zang Li, Yunsheng Pang, Hongyang Su, Jianchao Tu, Yunlong Wang, Bing Wen, Junzhang Zhu, Shaojie Zhu, Chengxiang Zhuo

High Energy Physics (Phenomenology)

Amplitude structure of $3\to 3$ scattering in a Mandelstam variable representation

We construct a dispersive representation of the relativistic $3\to 3$ scattering amplitude for three identical spinless particles in the $S$-wave. The two-particle subenergy, instead of the total energy, is used as the dispersive variable. This choice keeps the physical dispersive contour free of the kinematical cuts that complicate total-energy dispersion relations. By separating discontinuities across the two-particle subenergy cuts from that across the three-body cut, we derive a linear integral equation with one-particle exchange as the driving term. We further show that the solution satisfies three-body unitarity as a consequence of two-body unitarity, analyticity, and crossing symmetry. For pair-wise interactions, this representation can be rewritten into the form used for isobar-spectator scattering. Finally, we give prescriptions for contour deformation and the subtraction of poles in the two-body subsystem amplitudes, which continue the amplitude onto adjacent unphysical Riemann sheets and thus provide direct access to the analytic structure relevant for three-body resonance poles.
Read the full abstract

Computer Vision

Semi-Supervised Virtual Staining via Morphology Preservation and Histopathological Realism Constraints

Virtual staining aims to computationally generate target-stained histopathological images while reducing the cost and time associated with conventional staining procedures. However, existing methods rely predominantly on strictly paired and accurately registered training data, which are difficult and expensive to obtain in routine practice. To reduce this dependence, we propose a stable semi-supervised virtual staining framework that jointly exploits both limited paired data and abundant unpaired source images. Directly incorporating unpaired images is challenging because their generated results lack corresponding targets for supervision, potentially leading to unrealistic staining, morphological degradation, or even training collapse. To obtain reliable supervision from these images, Hessian-derived morphology preservation extracts structural cues from each source image and constrains the generated output to retain tissue morphology. Histopathological realism constraints further guide the output toward plausible target-stain characteristics, preventing the source-derived structural supervision from degenerating into contour enhancement or simple color transformation. Together, the two components suppress structural and appearance drift, stabilize semi-supervised stain translation, and promote the preservation of diagnostically relevant information. Extensive experiments on H&E-to-IHC translation for Ki67 and HER2, as well as FFPE-to-H&E translation, demonstrate consistent improvements in image quality, morphology preservation, robustness, and downstream diagnostic performance. Code will be available.
Read the full abstract

All authors Baoshun Wang, Weiping Lin, Linwu Wang, Yihuang Hu, Baptiste Magnier, Liansheng Wang

Chaotic Dynamics

Predicting Phase Ordering in Chaotic Maps and Coupled Map Lattices

Coupled logistic maps exhibit collective ordering of their directional phases. As the system parameter varies, the directional phases can undergo a transition from an in-phase state to an anti-phase state, while the individual map trajectories remain chaotic. In this work, we propose a data-driven machine learning (ML) framework based on parameter-aware reservoir computing (PARC) to predict order-parameter dynamics in two representative systems: a logistic map and a two-dimensional coupled map lattice (CML). For the logistic map, the reservoir is trained using only pre-crisis time series data at bifurcation parameter $\mu$ values below the attractor-merging crisis ($\mu_0 = 3.6786$). The trained reservoir reconstructs the full bifurcation diagram and correctly predicts the transition in the directional order parameter $M(\mu)$, from an ordered state ($M \approx 0$) to a disordered state ($M \neq 0$) across the crisis point. For the CML, we exploit the spatial homogeneity of the lattice: a single reservoir is trained on the dynamics of one representative lattice site and is then replicated across all $L^2$ sites during prediction, where $L$=50. The replicated reservoir correctly predicts the transition from in-phase synchronization ($\theta \approx 1$) to anti-phase clustered states ($\theta \approx 0$) at $\mu \approx 3.82$, where $\theta$ quantifies phase coherence across lattice sites.
Read the full abstract

Computation and Language

Disclosure-Gated User Simulation for Companion-Agent Evaluation

Playing the user with a large language model is standard in scalable evaluation, and has a repeatedly diagnosed failure: the simulated user is excessively cooperative, so a system under test can score by the number of questions it asks rather than by making the user willing to speak. We answer with a disclosure gate conditioning information release on the companion agent's behaviour: its state is a ladder of five ordered gates, merged onto three observable depth layers. We specify, ablate, and audit it. The gate is a load-bearing component of the environment: on a published companion-agent benchmark's English corpus, withdrawing per-example gate statements from training displaces the ranking of the 12 systems beyond the noise band of a seed re-run, while per-system scores show no detectable change. We state two acceptance criteria: a ranking must be order-preserving, and absolute scores must be scale-stable. Only one candidate passes both – the simulator we release – correlating at 0.993 with the published leaderboard. Prompting a frontier model instead barely moves the ranking while shifting every score up, invisible to ranking-only checks; we supply the specification, ablations, and downstream sensitivity analysis the benchmark's environment lacked.
Read the full abstract

All authors Yao Liu, Yu He, Jihao Huang, Xinyang Liu, Feng Zhou