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

Monday 14 September

High Energy Physics (Phenomenology)

Determination of charged pion unpolarised TMDPDFs from Drell-Yan measurements

We present an extraction of the unpolarised transverse momentum dependent (TMD) parton distribution functions (PDFs) for the valence quarks in the pion from experimental mea- surements of the Drell-Yan process. The procedure adopted in this project relies on the framework of TMD factorisation, and the inputs of nucleon TMDPDF, pion collinear PDF, as well as the Collins-Soper kernel. The non-perturbative behaviour of the TMDPDFs is parameterised using an ansatz function that contains three unknown parameters and implements an exponential decay at large transverse distances. This work employs the most recent results in perturbation theory, and utilises a precise estimation of the nucleon TMDPDF. We perform an investigation of data used in the fit. This investigation leads to important observations regarding the effects of the kinematics and binning methods in the experimental data. Our results agree qualitatively with the previous determinations from phenomenology and a lattice computation
Read the full abstract

All authors Wen-Chen Chang, Chia-Yu Hsieh, Chung-Wen Kao, C.-J. David Lin, Valentin Moos, Wayne Morris

Statistical Methodology

Coherence is not enough: Aggregation constraints across predictive distributions, forecasts and decisions

Forecasts made at different levels of aggregation are often required to agree—for example, regional forecasts should sum to the national total. Forecast reconciliation imposes such relationships, but the meaning of agreement depends on whether the constraint is applied to a predictive distribution, a reported summary such as a mean or quantile, or a decision based on the forecast. We show that these operations are not interchangeable. Even when every possible outcome from a predictive distribution satisfies an aggregation rule, its separately reported marginal quantiles need not add up; under nonlinear relationships, even the coordinatewise means may violate the constraint. We characterise when linear reconciliation preserves marginal quantiles and show that nonmedian quantiles generally cannot be preserved in a genuine hierarchy. We also explain why conditioning on an exact nonlinear constraint requires specifying how that constraint is observed or approximated, and identify when the predictive mean contains enough information for a constrained decision. In an application to Australian tourism data with 28 rolling forecast origins, the median discrepancy between two natural constructions of the national 95th percentile is 159.9% of the reconciled 90% prediction-interval width under bottom-up reconciliation. Rankings of mean- and quantile-based forecasts also reverse when squared-error loss is replaced by asymmetric inventory loss. These results provide a framework for deciding what should be made coherent and how competing methods should be evaluated.
Read the full abstract

Number Theory

On the largest prime factors less than $y$ of consecutive shifted primes

For an integer $n > 1$, let $P^+(n)$ be the largest prime factor of $n$, and let $P_y^+(n)$ denote the largest prime factor of $n$ not exceeding $y$. One of Erdős and Turán's conjectures asserts that the asymptotic density of integers $n$ satisfying $P^+(n)<P^+(n+1)$ is $1/2$. Furthermore, Wang conjectures that for $x\rightarrow\infty$, one has $\#\{p\leq x:P^+(p-1)<P^+(p+1)\}\sim \pi(x)/2$. In this paper, we show the following results. For any $3\leq y\ll x^{o(1)}$, for $x\rightarrow\infty$, we have \begin{align*} &\#\{n\leq x:P^+_y(n)<P^+_y(n+1)\}\sim \frac{1}{2}x, \&\#\{p\leq x:P^+_y(p-1)<P^+_y(p+1)\}\sim \frac{1}{2}\pi(x), \end{align*} where the first result extends a previous range $3\leq y\ll \exp(\log x/(100\log\log x))$ due to Rivat (2001). For any $0<\alpha<17/32$, let $y=x^\alpha$. Then there exists $h(\alpha)>0$ such that \begin{align*} \#\{p\leq x:P_y^+(p-1)<P_y^+(p+1)\}\geq(h(\alpha)+o(1))\pi(x). \end{align*} In particular, the function $h$ satisfies $\lim_{\alpha\rightarrow 0^+}h(\alpha)=1/2$. Similar result also holds for $\#\{n\leq x:P_y^+(n)<P_y^+(n+1)\}$, which improves a previous result $\lim_{\alpha\rightarrow 0^+} h(\alpha)=1/4$ due to Wang (2019).
Read the full abstract

Instrumentation and Methods for Astrophysics

FOSSIL: Thermo-mechanical architecture

FOSSIL (FTS fOr CMB Spectral diStortIon expLoration) is a proposed ESA M8 mission tailored to measure the spectral distortions of the Cosmic Microwave Background (CMB) with a sensitivity three orders of magnitude beyond the COBE/FIRAS legacy measurement. Achieving this sensitivity demands an extraordinarily challenging cryogenic architecture: the scientific instrument must be maintained at the lowest achievable temperature (4.5 K), while the detector focal plane assembly operates at 50 mK. This paper presents the thermal architecture of the FOSSIL payload and instrument, from the spacecraft service module at 293 K down to the sub-kelvin detector stage. The thermal design draws on heritage from the Planck and ARIEL missions and relies on a staged passive cooling chain comprising a multi-layer insulation blanket, three V-groove radiators (operating at approximately 130 K, 90 K, and 50 K), and a 25 K actively cooled shield fed by an ESA-provided 4 K cryocooler chain. Sub-kelvin temperatures are achieved via a multi-stage adiabatic demagnetisation refrigerator (ADR) system developed for NewAthena/X-IFU, providing continuous cooling at 1.8 K and 350 mK, and 50 mK with an 80% duty cycle. The Focal Plane Assembly (FPA), which houses four Kinetic Inductance Detector (KID) arrays at 50 mK, is thermally isolated from the 4.5~K bench via a carbon-fibre reinforced polymer (CFRP) hexapod structure with staged heat interception. We present the steady-state thermal budget across all stages, demonstrating comfortable margins at every temperature level, and discuss key thermal design drivers on the Blackbody Internal Reference (BBIR).
Read the full abstract

All authors Valentin Sauvage, Clémence de Jabrun, Sylvain Martin, Bruno Maffei, Nabila Aghanim, Anaïs Besnard, Bruno Borgo, Ivan Charles, Xavier Coulon, Sandrine Couturier, Francesco Cuttaia, Jean-Marc Duval, Morgane Loquet Le Gall, Giorgio Savini, Luca Terenzi, Berend Winter

Computation and Language

CITECHOICE: A Causal Audit of How Document Presentation Redistributes Citation Credit in Agentic Search

When several retrieved sources support the same claim, an answer engine cites some but not others. We call this decision citation allocation and introduce CITECHOICE, a causal audit of authentic multi-turn agentic search. From 129 everyday-query transcripts, CITECHOICE selects 113 same-call document pairs with independently verified support for the same pre-specified fact, without observing ranks or answer outcomes; blinded human review confirms 103. It runs a hash-verified 2-by-2 replay crossing pair order with jointly generated, fidelity-checked structured and prose renderings of one target while the rest of the transcript remains fixed. Three results emerge. First, and most importantly, structured rendering concentrates citation credit rather than clearly increasing source admission. It raises target citation count by +0.50 citations per answer (95 percent CI [+0.20, +0.84]; Holm-adjusted p=.033), without increasing total citations or reducing competitor credit. The pre-specified incidence effect (whether the target is cited at all) is +4.5 percentage points and inconclusive (95 percent CI [-1.4, +10.4]; p=.168). Second, observational position differences exceed controlled reordering effects: the citation-rate gap between rank 1 and rank 5 is 42.3 percentage points, compared with +7.9 percentage points in the main replay and 0.0 percentage points held out. Third, citation evaluation has a measurable noise floor. Although the aggregate count effect repeats under fresh decoding of 30 frozen families, 15 percent of binary decisions change and decoding accounts for an estimated 45 percent of single-generation family-effect variance. Together, these findings isolate what survives control: presentation can causally redistribute visible citation credit within frozen transcripts. They do not establish reliable source admission, a pure formatting mechanism, or a general rank advantage.
Read the full abstract

Algebraic Geometry

Noether's problem for $A_6$ and $A_7$

We prove that $B_kA_6$ is stably rational over every field $k$ in which $2$ and $-3$ are nonzero squares. Combining this with a theorem of Plans, we deduce that $B_kA_7$ is stably rational over every field $k$ of characteristic zero in which $2$ and $-3$ are squares.

Robotics

LieSpline-DP: Lie-Group B-Spline Diffusion Policy for Smooth Robot Manipulation

Diffusion Policy (DP) is a powerful Learning from Demonstration (LfD) method for robotic manipulation, yet it suffers from discontinuous and non-smooth trajectories. Spline-based action representations promote smooth motion within individual action chunks, but existing spline-based methods neither guarantee cross-chunk $C^2$ continuity nor account for the group structure of $\mathrm{SE}(3)$. We therefore propose LieSpline-DP, a Lie-group B-spline diffusion policy that generates end-effector trajectories directly on $\mathrm{SE}(3)$ and couples consecutive plans by sharing their boundary control poses, ensuring $C^2$ continuity throughout the entire planned trajectory. Across three real-robot tasks, LieSpline-DP produces lower trajectory jerk and higher task success rates than the DP baseline. The gains are particularly pronounced in real-world tasks involving liquids and flexible objects: in our real-robot experiments, LieSpline-DP achieved a 100% success rate on both pouring and bucket hooking, whereas the DP baseline achieved only 10% and 30%, respectively.
Read the full abstract

All authors Erxuan Xie, Bang Liu, Pingyun Nie, Xingkai Liu, Zhuang Fu, Bo Zhang

Computation and Language

EMR: Self-Evolving Medical Multi-Agent System via Experience Mining and Reuse

Large language model (LLM) driven multi-agent systems have shown promise in complex clinical reasoning, yet existing approaches rely on static strategies and lack persistent clinical memory, preventing self-evolving from prior diagnostic successes and failures. We present EMR, a self-evolving medical multi-agent system via Experience Mining and Reuse. EMR introduces a hierarchical clinical experience library that organizes accumulated knowledge into three levels: clinical principles, diagnostic patterns, and representative cases. During inference, EMR emulates multidisciplinary consultation: a planner agent coordinates domain-specific department agents for specialized reasoning, while a summary agent synthesizes their analyses into a final decision. Critically, EMR automatically extracts correct diagnostic insights and failure-related warnings from multi-agent reasoning trajectories, incrementally updating the experience library to guide future cases. Experiments on medical reasoning benchmarks demonstrate that EMR consistently outperforms state-of-the-art medical multi-agent baselines. Further analysis reveals that the hierarchical experience enables cross-specialty generalization and transfer across diverse LLM backbones, offering a scalable and in
Read the full abstract

All authors Dongsheng Shi, Yue Li, Xin Yi, Linlin Wang

Artificial Intelligence

T-LoopFormer: Token-Level Elastic-Depth Looped Transformers with Dynamic Routing

Looped Transformers have recently demonstrated strong performance in both reasoning and language tasks by reusing a shared set of parameters across multiple iterations, achieving parameter efficiency without sacrificing representational power. Besides, looped Transformers perform inference directly in the latent space (latent reasoning) to reduce the number of tokens consumed during inference, thereby achieving improved sample efficiency. However, these models typically apply a fixed recursion depth uniformly to every token, leading to suboptimal compute allocation and leaving significant efficiency gains on the table. In this work, we propose token-level elastic-depth looped transformers: T-LoopFormer. T-LoopFormer uses dynamic token-choice routing, which enables each token to adaptively determine its own number of loop iterations based on its hidden state, which could improve the token generation accuracy. Moreover, we further introduce recursion-wise KV cache, which maintains an independent key-value cache for each recursion loop of active tokens; this design ensures that tokens at different depths only attend to their corresponding cached states, effectively enabling faster autoregressive decoding. Extensive experiments show that T-LoopFormer achieves robust performance on a wide range of language modeling and zero-shot reasoning tasks, and our model could reach the lowest decoding latency, which validates the effectiveness of token-choice router and recursion-wise KV cache. Code is available at https://github.com/YuMingQian1234/T-LoopFormer
Read the full abstract

All authors Mingqian Yu, Wenpeng Zhang, Shaobo Cui, Peilin Zhao

Differential Geometry

Stability of Einstein 4-manifolds satisfying a chiral curvature condition

Let $(M,g)$ be a compact oriented Einstein four-manifold with Einstein constant $E$ and let $\widehat{R}^+$ denote the action of the Riemann curvature tensor on self-dual two-forms. We show that if $\widehat{R}^+ < 0$, then $g$ is strictly linearly stable for the Einstein-Hilbert functional, thus giving a chiral criterion for stability. Our result is stronger than the one by Fine-Krasnov-Singer, who conclude local rigidity from $\widehat{R}^+ < 0$ by proving stability for a different action functional. Our proof proceeds by showing that $(M,g)$ admits a natural spin$^h$ structure carrying a non-zero parallel spin$^h$-spinor. We then apply a lower bound on the Lichnerowicz Laplacian on traceless symmetric two-tensors in the presence of such a spinor.
Read the full abstract

Earth and Planetary Astrophysics

Spectral Evolution of Ceres' Surface and Implications for Space Weathering

The surface of Ceres exposed to the space environment will be gradually altered, but the evolution of its spectral properties remains poorly understood. Here we analyze visible and infrared spectra acquired by the Dawn mission to investigate spectral variations across different surface units. We identify local shifts of a few nm toward shorter wavelengths in the 2.7 $\mu$m band center, predominantly associated with freshly exposed slumping and impact materials. Our analysis suggests that compositional variations in Ceres' regolith are unlikely the explanation for the band center shift, and therefore it may represent a spectral evolution within 10$^6$ years. Combining previous studies about the spectral reddening on Ceres, we propose a two-stage spectral evolution on Ceres. The blue-shifted regions are interpreted as the youngest materials. Within 10$^6$ years after exposure, these materials become spectrally bluer, the 2.7 $\mu$m band weakens, and its center shifts toward longer wavelengths. Then the evolution is characterized by spectral reddening and strengthening of the 2.7 $\mu$m band without the band center shift. Based on the previous laboratory work and observations on other C-complex asteroids, we suggest that the observed spectral variations may be attributed to the chemical and/or physical changes of the regolith, possibly associated with space weathering.
Read the full abstract

All authors Qing Zhang, Jian-Yang Li, Yazhou Yang, Minge Liu, Yang Liu

Probability

Spectral Gap Bounds for Langevin Dynamics in Mixed Spherical Spin Glasses

We prove lower bounds on the relaxation time $1/\gamma_{N,\beta}$ of Langevin dynamics for mixed spherical spin glasses with even mixture $\xi(x)=\sum_{p\ge4}\gamma_p^2x^p$, under a one-step-replica-symmetry-breaking-type standing assumption of strict threshold separation $E_0(\xi)>E_1(\xi)>E_2(\xi)$; the pure spherical $p$-spin glass with even $p\ge4$, for which the assumption is a theorem, is recovered as a corollary. The main result is an aggregate Eyring-Kramers bound whose exponent sums conductances over the exponentially many index-one saddles above the lowest saddle level $-NE_1(\xi)$, combining the saddle complexity $\Theta_{1,\xi}$ with a half-determinant Hessian statistic – an entropic contribution that the classical single-saddle picture misses. The single new random-matrix ingredient of the mixed model is that the conditional Hessian at a critical point is a randomly shifted GOE matrix: the radial derivative is no longer determined by the energy (Euler's identity degenerates exactly in the pure case), and every landscape rate becomes a one-dimensional supremum over the scalar shift with a Gaussian penalty. At low temperature the aggregate exponent exceeds the unconditional free-energy bound by $\frac12\log(\beta e)-\Xi^{\mathrm{EK}}_{1,\xi}(-E_1(\xi))-C_{\xi,b}\beta^{-1/2}$ with $-\Xi^{\mathrm{EK}}_{1,\xi}(-E_1(\xi))\ge\frac14\log\xi''(1)+\frac14>0$, so the refinement is a strict improvement for all sufficiently large fixed $\beta$; the onset temperature and constants $b_*(\xi)$ and $C_{\xi,b}$ produced by the proof are not numerically explicit. Two companion results – a sequential Arrhenius bound with the explicit constant $E_0(\xi)-E_1(\xi)$ and a fixed-temperature free-energy bound – come with complete, self-contained proofs. All bounds are one-sided; identifying the mechanism the dynamics actually realizes would require a matching upper bound, which remains open.
Read the full abstract

General Relativity and Quantum Cosmology

Thermodynamic stability of the new black hole solution in $f(R,G,T)$ gravity

In this literature, we derive an approximate black hole metric within the \( f(R, G, T) \) gravity framework, where the black hole is surrounded by an anisotropic fluid as the matter source. This approach allows us to study the influence of anisotropic pressures on the black hole structure. We then investigate the thermodynamic properties of the approximate solution, considering various types of matter fields. We analyze the thermodynamic quantities such as Wald's entropy, Hawking temperature, and specific heat for each case. After examining the specific heat, we find that the black holes are thermodynamically stable, with positive specific heat values indicating stability. This result suggests that the black holes under these conditions do not undergo thermal instability, ensuring their long-term viability.
Read the full abstract

Classical Analysis and ODEs

Invariant sets of the double Hilbert transform

We present an example that shows that the double Hilbert transform $\H$ can have the characteristic function of an open set of finite measure in $\mathbb{R} \times \mathbb{R}$ as an approximate eigenvector for the eigenvalue $1$. The same holds for the dyadic model of $\H$, the double dyadic shift. We further construct non-diagonal sets whose indicators are fixed by $\H$ – strips bent along the Boole curve $x=y-1/y$ and, more generally, along level curves of real Herglotz functions with purely singular measures – and study the class of invariant sets which they generate. At positive density we construct an example built on three families of parallel lines, related to the carry cocycle $\{u\}+\{v\}-\{u+v\}$, and the fans – alternate sectors of an odd number of concurrent lines – which are the simplest members of the invariant cones, a class that we describe completely. We then study the classification of the invariant sets whose boundary lies on countably many lines. Its combinatorial half is settled completely: a locally finite arrangement of lines in which no point lies on exactly two lines is a family of parallel lines, a pencil of concurrent lines, or an affine image of the triangular lattice. The classification is further connected, via projective duality, with configurations of Sylvester–Gallai type, and, via a change of variables, with two-valued superpositions of traveling waves.
Read the full abstract

All authors Evgeny Abakumov, Komla Domelevo, Stefanie Petermichl, Alexei Poltoratski

High Energy Physics (Phenomenology)

Precision Sum Rule for Nucleon Isovector Polarizabilities and the Proton-Neutron Mass Difference

The precision of the electromagnetic proton-neutron mass difference $\delta m_\mathrm{QED}$ extracted from the Cottingham formula hinges on a subtraction function whose low-energy normalization $\bar S(0)$ is fixed by the isovector combination of the proton and neutron polarizabilities, $(\alpha_{E1}-\beta_{M1})^{p-n}$. We derive a dispersive sum rule for this combination in terms of $s$-channel photoabsorption cross sections and the product of $t$-channel $\gamma\gamma\to\pi\eta/K\bar K_{I_t=1}$ and $\pi\eta/K\bar K_{I_t=1}\to N\bar N$ amplitudes, without invoking Reggeon dominance. Combining empirical pion-photoproduction multipoles with coupled-channel Muskhelishvili-Omnès representations of the scalar-isovector $\pi\eta/K\bar K$ amplitudes, we obtain $(\alpha_{E1}-\beta_{M1})^{p-n}=-0.82(33)\times10^{-4}\,\mathrm{fm}^3$, fixing its sign and reducing the uncertainty by one order of magnitude compared with the previously known value. This result yields $\bar S(0)=-0.56(28)\,\mathrm{GeV}^{-2}$, leading to $\delta m_\mathrm{QED}=0.75^{+0.02}_{-0.04}\,\mathrm{MeV}$, substantially more precise than previous Cottingham determinations. The negative $\bar S(0)$ also provides a low-energy test of Reggeon dominance in the subtraction function.
Read the full abstract

Signal Processing

Emulation of Closely Spaced Spatial Sensing Targets Using a Parabolic Cylindrical Compact Range with a Multi-Feed Amplitude-Phase Controlled Linear Array

Integrated sensing and communication (ISAC) base stations (BSs) are emerging as critical infrastructure for 6G networks, requiring advanced over-the-air (OTA) testing methodologies that can emulate dynamic, multi-target scenarios under far field (FF) conditions. However, existing testing solutions often suffer from limited angular flexibility, insufficient bandwidth, and prohibitive costs. {To address these challenges, this paper proposes an OTA testing system that integrates a radar target simulator (RTS), a multi-port amplitude and phase modulation (APM) network, and a multi-feed linear array with a single parabolic cylindrical compact range (SPCCR).} The proposed system is capable of emulating closely spaced spatial targets with varying incident azimuth angles. This setup meets the essential conditions for ISAC BS testing, including FF characteristics, wideband and multi-band compatibility, and the capability for sensing multiple dynamic closed-spaced targets. The feasibility of the proposed method is validated through both FEKO simulation and measurement. Experimental results demonstrate that the system successfully emulates multiple closely spaced targets within an azimuth range of $\pm 10^\circ$ with high emulation accuracy while maintaining excellent quiet zone (QZ) performance.
Read the full abstract

All authors Zhangzhang Jiang, Guokai Jiang, Le Yu, Chunhui Li, Zhengpeng Wang, Wei Fan

Computer Vision

PACE: Progressive Angular-to-Norm Contrastive Embedding

Multimodal embedding models encode heterogeneous inputs into a shared embedding space, enabling efficient similarity computation across modalities and tasks. Most existing methods optimize cosine-based contrastive objectives, which promote stable training but restrict semantic compatibility to angular geometry, precluding embedding norms from serving as an additional semantic signal. However, directly optimizing the more expressive dot-product similarity, which leverages both angular and norm information, underperforms cosine-based training and exhibits unstable training dynamics. We attribute this discrepancy to premature optimization-space expansion, manifested as angular–norm entanglement and directional anisotropy in the representation space and further compounded by full-parameter fine-tuning. In this paper, we propose PACE, a two-stage framework that progressively expands both the representation and trainable parameter spaces. Stage I combines cosine-based objective with low-rank adaptation to establish a reliable angular geometry within constrained optimization spaces. Stage II switches to dot-product similarity and full-parameter fine-tuning, enabling embedding directions and norms to jointly encode semantic information. We further introduce Focal Embedding Loss, a confidence-adaptive objective that downweights queries with high positive retrieval confidence while emphasizing ambiguous queries with competitive negatives. Experiments across multiple backbone scales and diverse multimodal embedding tasks consistently validate the effectiveness of PACE.
Read the full abstract

All authors Yanping Li, Wei Zhou, Yawen Liu, Yibo Wang, Ke Zhu, Guangda Huzhang, Qing-Guo Chen, Zhao Xu, Jun Zhang, Wei Wei

Materials Science

Evaluating Predicted Densities, Hamiltonians, and Density Matrices as Periodic SCF Initializers

Learned electronic states are usually evaluated by prediction error, even though their intended use is to accelerate the self-consistent-field (SCF) loop of density functional theory (DFT). We ask whether lower offline error actually yields a better SCF initializer. We construct $\rho$HD-43K, a 43,851-crystal DFT corpus with aligned charge-density, Hamiltonian, and density-matrix labels, and extend the solver workflow to inject all three predicted states. The closed-loop benchmark compares direct converged-state prediction against residual prediction from solver-native references on a frozen test set of non-magnetic crystals, using paired convergence, iteration, and SCF-loop timing measurements. Injecting the exact converged density matrix or Hamiltonian as an oracle upper bound cuts the median SCF count from 16 to one, revealing substantial acceleration headroom. Among learned inputs, direct charge-density prediction (Charge3Net-E3) accelerates about 91% of paired crystals, saves a median of three SCF iterations, and yields a $1.18\times$ SCF-loop speedup. In contrast, the tested residual-density and matrix initializers do not consistently improve over the standard superposition-of-atomic-densities baseline. These results establish that target-space accuracy alone is insufficient: a learned initializer must also be compatible with the nonlinear solver trajectory, and its utility must be measured in the loop.
Read the full abstract

Computer Vision

Weakly Supervised Spatial Grounding for Discriminative Attention-Based Ultrasound-Histopathology Alignment in Prostate Cancer Grading

Unpaired cross-modal distillation transfers grade structure from histopathology into a micro-ultrasound (micro-US) encoder by aligning a pooled needle-region embedding to a frozen histopathology teacher under grade-group correspondence alone. A single objective is thereby required to serve two distinct functions: rendering patch features discriminative of tissue state, and selecting which patches enter the pooled representation. We decouple them. Weak spatial supervision derived from percentage involvement, recorded routinely at biopsy, constrains the predicted proportion of malignant tissue within each core, acting on the encoder features independently of the alignment objective. The alignment loss then operates on features that differ across a core, and attention concentrates on a subset of patches rather than remaining near-uniform. On 7,166 biopsy cores from 811 patients across seven centers under patient-level 5-fold cross-validation, the method reaches 67.1 macro AUC and 68.5 csPCa AUC, against 61.2 and 52.8 for the existing unpaired alignment method and 63.1 and 62.6 for the strongest unimodal baselines. Ablation against existing attention regularizers designed to prevent attention-uniformity collapse shows that such regularizers do not substitute for label-derived supervision: they constrain the attention distribution, whereas the signal required acts on the features that attention reads.
Read the full abstract

All authors Obed Korshie Dzikunu, Emma Willis, Mohammad Mahdi Abootorabi, Mohamed Harmanani, Zhuoxin Guo, Ferdinand Luger, Adam Kinnaird, Brian Wodlinger, Parvin Mousavi, Purang Abolmaesumi

Machine Learning

Multi-source Transfer Learning of Time Series with a Shapelet-based Distance Measure

Transfer learning is an effective technique for addressing data scarcity in deep learning for time series classification, but its success depends on the selection of source datasets. Conventional transferability estimation methods are often computationally expensive, as they require fully pre-training a model on each potential source dataset to assess its suitability. This paper introduces a novel, training-free source selection method named Shapelet Matching. Our approach first identifies discriminative shapelets from the target and potential source datasets. Then, Shapelet Matching quantifies dataset similarity by comparing the extracted sets of shapelets. To mitigate the risk of negative transfer from selecting an unsuitable single source, we introduce a multi-source transfer learning method. We select several source datasets based on their shapelet-based similarity scores, combine them into a single multi-source dataset, and use this aggregated dataset for pre-training. The model is then fine-tuned on the target task. We evaluated our method on 128 datasets from the UCR Archive using both temporal CNN and Transformer architectures. The empirical results demonstrate that our multi-source pre-training reduces the risk of negative transfer on average. Shapelet Matching achieves the strongest performance for the CNN backbone and remains competitive for patch-based Transformer architectures, while avoiding the cost of pre-training a separate model for every candidate source.
Read the full abstract

Differential Geometry

K-contact manifolds admitting some geometric solitons with Semi-Symmetric Non-Metric Connection

In this paper, we introduce some type vector fields with respect to a semi-symmetric non-metric (SSNM) connection. We investigate several geometric properties of a K-contact manifold equipped with an SSNM connection and provide a concrete example to justify the relation between the scalar curvature of the SSNM connection and Levi-Civita connection that we have obtained in this paper. Furthermore, we have found the nature of Riemann solitons, conformal Ricci solitons and conformal $\eta$-Ricci-Yamabe solitons on K-contact manifolds admitting a SSNM connection.\\ Finally, we determine the necessary and sufficient conditions for such a manifold to be $\Tilde{\tau}$-semi-symmetric, quasi-conformal-semi-symmetric and pseudo-projective-semi-symmetric.
Read the full abstract

Combinatorics

A lower bound on the density of prefixes with maximal palindromic length

The palindromic length $PL(u)$ of a nonempty finite word $u$ is the least number of nonempty palindromes whose concatenation is $u$. For an infinite word $w$, let $P_w(n)$ be the number of nonempty palindromic prefixes of $w[1,n]$, and let $T_w(n)$ consist of those prefixes $w[1,j]$, $1\leq j\leq n$, whose palindromic length equals the maximum attained among the nonempty prefixes of $w[1,j]$. We prove the finite inequality $|T_w(n)|\geq P_w(n)$ for every $n\geq1$. In particular, if $w$ has infinitely many palindromic prefixes, then \[ \liminf_{n\to\infty}\frac{|T_w(n)|}{P_w(n)}\geq1. \] The coefficient $1$ is optimal, already for a nonconstant periodic word. The proof uses chains of occurrences connected by palindromic factors, together with trimming and reflection arguments that preserve lower bounds on palindromic length.
Read the full abstract

Computation and Language

Improving Mathematical Reasoning Capabilities in Large Language Models via Reasoning Process Error Classification

The reasoning ability of large language models (LLMs) is a critical factor for practical LLM-based applications. To investigate the current reasoning capability of LLMs, we clarify the types of errors that arise in LLMs' reasoning processes on mathematical datasets. We focus on problems where LLMs produce an incorrect answer. We define errors in the reasoning process as reasoning errors and manually analyze the features of reasoning errors. We defined and classified 21 error classes and identified the frequently occurring classes among them. Beyond qualitative evaluation, we leverage the evaluation results to improve the reasoning capability. We designed a prompt that explicitly focuses on eight error classes. The experiments demonstrate that this prompt effectively improves reasoning performance. Furthermore, the results suggest that the frequent reasoning errors identified in this paper are common across LLMs of comparable scale.
Read the full abstract

Computer Vision

Automated Perceptually-Motivated Assessment of Photographic Consistency in Paired Clinical Photographs: Pipeline Development and Internal Evaluation

Purpose: Paired pre- and post-operative photographs are the standard unit of evidence for plastic surgical outcomes, yet no objective metric verifies whether two images of the same patient were captured under conditions consistent for comparison. Approach: We developed a perceptually motivated pipeline that analyzes pre/post pairs across thirteen calibrated sub-metrics, partitioned by unsupervised correlation-structure analysis into five data-driven clusters (photometric, texture / sharpness, pose, illumination direction, and pitch), averaged within each cluster and combined across clusters by a weighted sum into a single consistency score. Each sub-metric is calibrated so that its median difference across published within-patient pairs scores 0.5, which is a reference point and carries no pass/fail meaning. The pipeline was calibrated on 134 matched within-patient published pre/post pairs and evaluated against identical-image pairs, synthetic-perturbation pairs, and 134 mismatched cross-publication pairs. Results: The master consistency score S separated matched from mismatched pairs (sensitivity index d' = 2.15, 95% confidence interval (CI) [1.83, 2.55]; area under the receiver operating characteristic curve AUC = 0.928, 95% CI [0.896, 0.959]), closely matching Gaussian-equal-variance predictions. The three head-pose angles did not fall in one cluster: yaw and roll grouped together while pitch separated. Identical pairs scored at ceiling (S = 0.99) and the master score fell monotonically with perturbation magnitude on all five perturbation axes. Conclusions: The score quantifies photographic comparability, not aesthetic or surgical quality, and provides a freely available web tool for auditing the photographic comparability of pre/post pairs, pending validation against expert judgment.
Read the full abstract

All authors Derrick Lin, Samantha Rabinovich, Joclin Rabinovich, Kassra Garoosi, Sumun Khetpal, Evan Delanoy, Neel Bhardwaj, Jason Roostaeian

Information Retrieval

Top-K Is Not a Budget for Hybrid Retrieval

Modern hybrid retrieval for RAG typically fuses the Top-$L$ results from dense and sparse retrievers, but a fixed truncation depth may not transfer across changing queries and corpora. Exact fusion removes the dependence on a fixed depth, yet completing a specified Top-$K$ still incurs variable access costs. We present DiBud, which takes an access budget directly as input and incrementally certifies and returns an exact prefix of the RRF ranking over the full lists. Selective access increases certified output within the budget, while budgeted stopping bounds accesses per request. Experiments on five query sets reveal long-tailed costs for completing exact Top-20. At a budget of 2048 accesses, DiBud increases mean certified output within the first 100 positions by 7.86% over balanced access. After budget calibration for 95% quality retention, held-out queries retain 95.05%–97.68% of mean nDCG@20 while using 65.92%–99.53% fewer accesses than completing exact Top-20.
Read the full abstract

Robotics

C$^2$Nav: Compare Before You Commit for Zero-Shot Vision-and-Language Navigation

Zero-shot vision-and-language navigation in continuous environments (VLN-CE) increasingly places foundation vision-language models (VLMs) inside the navigation loop. Existing systems commonly request cardinal outputs such as waypoints, pixels, headings, progress values, or absolute arrival decisions, coupling a generative response to geometric magnitude or an irreversible commitment. We study a complementary model-robot interface: the VLM compares controller-constructed alternatives, while geometry, thresholds, action magnitude, and execution remain on the physical side. We instantiate this idea in C2Nav, a training-free framework with three coordinated faculties. Seeing performs ordinal Gaze Election over physically vetted candidate views; Remembering maintains a compact route sketch and compares adjacent instruction-leg hypotheses; and Arriving combines a hesitation ladder, look-back comparison, and revocable walk-back for reliable stopping. On the public OpenNav R2R-CE 100 protocol, C2Nav with Qwen3-VL-8B-Instruct obtains 41.0% OSR, 31.0% SR, and 16.7% SPL, while the same interface with the standard GPT-5.5 model reaches 54.0% OSR, 44.0% SR, and 29.0% SPL. Whole-faculty ablations reduce SR to 14.0% without Seeing, 25.0% without Remembering, and 29.0% without Arriving. Matched role inversions that replace only the comparative answer form with cardinal/absolute questions reduce SR to 12.0%, 28.0%, and 21.0% in the spatial, transition, and terminal slots, respectively. The results indicate that a constrained decision interface and stronger VLM reasoning are complementary rather than interchangeable.
Read the full abstract

Analysis of PDEs

Normalized solutions to the NLS equation in the ball for any prescribed mass

Given $\rho>0$, we consider the problem \[ \text{find $(\lambda,u) \in \mathbb{R} \times H_0^1(B)$ such that } \begin{cases} -\Delta u+\lambda u = |u|^{p-1}u & \text{in } B \\ \int_B u^2\,dx = \rho, \end{cases} \] where $B$ is a ball in $\mathbb{R}^N$, $N \ge 1$, and $1<p<2^*-1$. Without any further restriction on $N$, $\rho$ and $p$, we prove the existence of infinitely many radial solutions, and, in dimension $N \ge 4$, of at least one non-radial solution.
Read the full abstract

General Relativity and Quantum Cosmology

Dynamical Structure of Einasto Dark Matter Wormholes: Thin-Shell Stability and Particle Transitions

In this manuscript, we develop topologically charged static and spherically symmetric wormhole (WH) configurations in Einstein gravity by employing the Einasto dark matter profile to construct an analytic shape function. The resulting spacetime satisfies the fundamental traversability conditions, while the matter sector exhibits localized violations of energy conditions near the throat, indicating confined exotic matter content. We analyze equilibrium through the TOV framework, revealing a non-equilibrium force competition in which hydrostatic and anisotropic contributions self-organize to sustain static configurations within specific parameter domains. Stability is examined via the anisotropy parameter and further extended to linearized radial perturbations of the surrounding shell, uncovering parameter-dependent stability windows that characterize the system's nonlinear response. A detailed dynamical analysis shows that the monopole charge parameter significantly restructures the effective potential landscape, modifying curvature and inducing qualitative transitions in particle motion. The coupled interplay between monopole charge, angular momentum, and particle energy governs a nonlinear transition from tightly wound quasi-bound states to unbounded scattering trajectories. This transition reflects an emergent dynamical phase structure in the WH spacetime. Furthermore, the complexity factor exhibits strong localization near the throat and vanishes asymptotically, indicating that structural complexity is confined to the inner nonlinear regime, whereas a volume integral quantifier is employed to estimate the total exotic matter content required to sustain the configuration.
Read the full abstract

All authors M. Yousaf, Faisal Javed, Gulzoda Rakhimova, Farkhod Botirov, Farruh Atamurotov

Quantum Physics

A train–prune–readout–rewrite workflow for interpretable quantum learning

AI for Science aims not only to predict complex physical systems from data, but also to extract mathematical structure and physically testable representations from learned models. Here, a train–prune–readout–rewrite workflow is developed that separates physical-domain grounding from three increasingly stringent analysis claims: algebraically equivalent readout of a trained predictor, compact teacher-faithful symbolic rewriting on the sampled physical domain, and transformation-based tests of learned internal representations. The workflow is implemented with complex-valued Kolmogorov–Arnold networks, whose explicit edge functions enable post-pruning analytic readout of the retained computation. In analytically controlled single-qubit tasks, rewriting recovered the quadratic structure of purity, whereas von Neumann entropy yielded only a domain-bounded symbolic surrogate; physics-aligned variable grouping preserved symbolic fidelity. For two-qubit entanglement-related tasks, shared learning exposed a common internal representation whose physical content was interrogated directly. Local-unitary transformations rejected a direct invariant-coordinate interpretation, while fixed-decoder transfer showed that the shared activation carries Pauli-correlation information in a transformation-consistent form. Task-related invariant spectral features were subsequently recovered through low-order nonlinear readouts. Separate predictive tests retained high accuracy for three-qubit classification and controlled ten-qubit purity regression with over one million complex inputs. These results establish an evidence-resolved framework for distinguishing physical grounding, readable computation, faithful symbolic compression and transformation-tested physical structure in constrained complex-valued scientific learning.
Read the full abstract

All authors Siran Zhang, Shuming Cheng, Xiang Li, Jinyi Liu