Optimization and Control
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Thursday 3 September
We study optimizing $L$-smooth and $\mu$-Polyak-Lojasiewicz (PL) objectives with unbiased stochastic gradients under $\alpha$-heavy-tailed noise ($1<\alpha\le 2$). In the unrestricted high-dimensional setting, we establish the noise-adaptive lower bound for attaining $\epsilon$-suboptimal function value. Under the generalized mirror-PL condition, we propose a centered-clipped mirror-descent method that achieves a high-probability upper bound matching the lower bound up to logarithmic factors. We further provide the improved tight stochastic complexity in fixed dimensions.
Mesoscale and Nanoscale Physics
Strongly anisotropic non-Kramers electron spin as a quantum coherence probe of angular fluctuations
Strongly anisotropic non-Kramers rare-earth ions combine giant longitudinal g-factors with a vanishing transverse component imposed by time-reversal symmetry, a combination that makes their spin transitions exquisitely sensitive to the orientation of the applied magnetic field. We show that this sensitivity carries a dual identity: it is simultaneously an overlooked decoherence channel and the basis for a spin-coherence-based angular probe. Using pulsed electron paramagnetic resonance at X-band, we report the first measurements of the quantum coherence of Tb$^{3+}$ in a native-doped CaWO$_4$ crystal (15 ppb) and map the Hahn-echo coherence time $T_2$ as a function of temperature (2 to 10 K) and resonant field ($10^3$ to $10^4$ G). A parameter-free model combining spin-lattice relaxation, instantaneous diffusion and spectral diffusion from all independently quantified impurities overestimates $T_2$ by an order of magnitude at low temperature and wrongly predicts the field dependence of $T_2$, inconsistent with the observed monotonic decrease of $T_2$ with $B_r$. A two-parameter extension, including dynamical angular fluctuations of the crystal axis, reproduces the full dataset across multiple setups and laboratories. Two controlled experiments nominally identical except for different mechanical configuration of the setup establish the mechanical origin of the dominant contribution. The two-parameter extension corresponds to an angular amplitude noise spectral density of overall order 36 n{\deg}/$\sqrt{Hz}$ from global external vibrations (ranging from 10 to 66 n{\deg}/$\sqrt{Hz}$ depending on the exact setup mechanical configuration) estimated at $\sim$ 2.5 kHz plus a temperature-dependent contribution assumed to come from local phonon-driven angular jitter. It identifies and highlights a decoherence pathway of practical relevance to any anisotropic solid-state spin system.
Read the full abstract
Functional Analysis
A $p$-summability approach to the weak maximizing property
Motivated by the weak maximizing property ($\mathrm{WMP}$), we investigate $p$-summability conditions on maximizing sequences of bounded linear operators. Since the naive $p$-summability formulation is independent of $p$ and collapses to norm attainment for all operators, we introduce the weakly $p$-singular maximizing property ($\mathrm{SMP}_p$), based on maximizing sequences with no weakly $p$-summable subsequence. We completely characterize when the pairs $(\ell_p,\ell_q)$ and $(\ell_p,c_0)$ have the $\mathrm{SMP}_r$, revealing sharp contrasts with the $\mathrm{WMP}$. We also characterize the Schur property of order $p$ via the $\mathrm{WMP}$ and the $\mathrm{SMP}_r$. Under relative weak $p$-precompactness and suitable Dunford-Pettis-type assumptions, we characterize universal norm attainment for operators from $X$ to $Y$ in terms of either the $\mathrm{SMP}_q$ for adjoint operators or the weak$^{*}$-to-weak$^{*}$ maximizing property, and derive corresponding duality consequences for the $\mathrm{SMP}_p$. Finally, we introduce $p$-convergent perturbation properties for operators and their adjoints, characterize them for classical sequence spaces, and show that the $p$-convergent perturbation property is strictly weaker than the $\mathrm{SMP}_p$.
Read the full abstract
Neurons and Cognition
High-Order Triadic Functional Connectivity in the Brain and Beyond
Here, we report high-order functional network connectivity as a promising way for studying the brain connectome. Traditional functional connectivity approaches capture only pairwise relationships between brain regions, overlooking complex multivariate dependencies that underlie cognition and behavior. First, we demonstrated that high-order interactions capture more information and can distinguish between resting-state and task-state brain activity. Second, we introduce a matrix-based entropy-functional method for estimating triadic interactions, which are statistical dependencies among triplets of brain regions, and apply it to large-scale functional brain networks. The resulting triadic networks revealed distinct community patterns that complement those observed in traditional pairwise functional connectivity analyses and simultaneously capture additional connection information. Despite the potential combinatorial explosion of triadic configurations, the networks exhibited constrained and hierarchical structures that allowed computation and interpretation. These findings position triadic connectivity as a promising next-step functional connectivity framework for probing brain network organization and high-order neural interactions, while also highlighting key biological and technical challenges that require careful consideration.
Read the full abstract
Signal Processing
Low-Complexity Sequential Detection Framework for Single-Channel Co-Frequency Signal Separation
Practical separation of single-channel co-frequency signals (SCCFSs) is hindered by the prohibitive computational complexity of benchmark algorithms. To address this issue, we propose a low-complexity separation framework based on sequential detection (SD), in which signal separation is cast as a sequential path search over a trellis. To support both hard-decision detection and log-likelihood ratio (LLR) extraction, we develop two algorithms within this framework: the SD-based separation (SDS) algorithm and its soft-output variant (SO-SDS). Furthermore, SDS employs a windowing strategy combined with dynamic pruning to concentrate computational resources on high-probability paths, thereby enabling efficient detection of transmitted symbol sequences. Building upon SDS, SO-SDS further incorporates a state completeness verification mechanism (SCVM) to estimate bit LLRs, thus facilitating subsequent soft decoding. Numerical results show that, compared to benchmark algorithms, SDS achieves significant complexity reduction without degrading separation performance, while SO-SDS offers notable computational savings with only modest LLR accuracy loss. Notably, the computational complexity advantage of the proposed algorithms over benchmark algorithms grows substantially with increasing modulation order.
Read the full abstract
Computer Vision
IchthyoNoma: Nomenclature and Context Sensitivity of Zero-Shot Biological Vision–Language Models for Bangladeshi Freshwater Fish Recognition
Zero-shot vision-language models (VLMs) are increasingly used as training-free species recognizers, but reported accuracy can reflect more than visual species knowledge. We audit CLIP, BioCLIP, BioCLIP2, and a multilingual Jina CLIP v2 control on seven freshwater-fish categories from two Bangladeshi sources (10,321 images). BioCLIP2 reaches 72.36% on BFF-15 with English common names and 68.91% on SylFishBD with scientific names, versus 25.15% and 14.40% for generic CLIP. BioCLIP2 Bengali prompts are near chance in balanced accuracy (14.22-14.29%); Jina partially recovers Bengali discrimination to 21.89% and 16.36%, but bare Bengali names return to 14.29% on both sources. Paired SylFishBD interventions show no significant weak-blur effect, modest losses from stronger blur/gray masking, a larger white-mask artifact, and strong species dependence. Zero-shot biological VLM scores therefore jointly reflect biological specialization, multilingual alignment, nomenclature, prompt formulation, and context.
Read the full abstract
Robotics
MulDP: Multimodal Diffusion Policy for Autonomous Quadruped Parkour Navigation across Complex Terrains
Quadruped robots have demonstrated impressive agility in parkour locomotion across complex terrains. However, most systems still rely on human intervention for high-level planning, and autonomous parkour navigation remains underexplored. The key challenges include fine-grained velocity regulation, long-horizon anticipatory behaviors, and tight coupling between perception and embodied execution. To address these challenges, we propose a Multimodal Diffusion Policy (MulDP) that integrates visual perception with robot proprioception and goal information to generate temporally coherent and anticipatory navigation velocity commands, tightly coupling perception with embodied control to enable robust autonomous navigation. To support the training of MulDP, we construct the first Quadruped Parkour Navigation Dataset (QPND), a multimodal dataset that encompasses diverse navigation behaviors and complex terrains. Extensive simulation and real-world experiments demonstrate that MulDP enables robust long-horizon autonomous navigation and effective traversal across complex terrains.
Read the full abstract
General Relativity and Quantum Cosmology
Black holes in alternative theories of gravity
Black holes, with their strong gravitational fields, provide an important testing ground for theories of gravity beyond General Relativity. Among the many proposed alternatives, considerable recent work has focused on scalar-tensor theories in which a scalar field couples to higher-curvature terms. The black hole solutions that arise in such theories can differ significantly from the Schwarzschild and Kerr solutions of General Relativity. Characteristic properties of these black holes include instabilities, shadows, and gravitational wave spectra, which can be used to constrain the couplings of the underlying theories.
Read the full abstract
Audio and Speech Processing
Fairness Evaluation of Edge-AI Implementation for Cleft Lip and Palate Speech ASR
Automatic speech recognition (ASR) remains challenging for individuals with cleft lip and palate (CLP) because of limited pathological speech data and large variations in speech characteristics across speakers and severity levels. These recognition difficulties can reduce the accessibility of voice-based human-computer interaction, particularly when cloud-based ASR services are unavailable or unreliable. This work investigates a severity-aware and edge-deployable ASR framework for improving recognition of CLP speech using Whisper-small. The model was fine-tuned using different combinations of normal and CLP speech representing mild, moderate, and severe conditions, together with a CLP-only training configuration, to examine how the inclusion of different severity levels influences recognition performance and fairness across speakers. The pretrained model produced pooled word error rate (WER) and phoneme error rate (PER) values of 62.46% and 52.72%, respectively. Severity-aware fine-tuning substantially improved performance, reducing the best pooled WER to 22.72% and the best pooled PER to 18.44%. Training with a broader representation of CLP severity levels also provided the best overall balance between recognition accuracy and performance consistency across severity groups. Deployment on an NVIDIA Jetson platform demonstrated real-time inference for all fine-tuned models, with real-time factors of 0.167-0.171 and peak GPU memory usage of approximately 566 MB. The results demonstrate that incorporating severity diversity during ASR adaptation can substantially improve recognition of CLP speech while reducing performance disparities across severity groups. The proposed approach further enables low-latency, Internet-independent speech interaction on edge devices, supporting more accessible and inclusive voice-based human-computer interaction for individuals with CLP.
Read the full abstract
Computer Vision
Sharpening the Ensemble: An SSIM-Aligned Residual Refiner for Brain-MRI Inpainting Post-Processing
Brain-MRI inpainting replaces a masked region of a scan with synthesized, anatomically plausible healthy tissue, so that analysis tools built for healthy brains can be applied to images they would otherwise reject. On the BraTS local-synthesis benchmark, which ranks submissions on the structural similarity index (SSIM), the peak signal-to-noise ratio, and the mean squared error (MSE) jointly, the strongest recent models are accurate, but several report blurry synthesized regions and attribute this to the mean-seeking behavior of the $\ell_1$ and MSE terms in their training losses. We address this in post-processing, forming a deep ensemble of the two co-first-place 2025 models and training a lightweight residual refiner on the ensemble's own outputs under an $\ell_1$ loss augmented with a structural-similarity term whose weight $\lambda$ we vary. At a moderate $\lambda$ the refiner improves SSIM over the ensemble, from $0.8767$ to $0.8780$ on a held-out reproduction of the official scorer and from $0.8555$ to $0.8572$ on the official validation leaderboard, with essentially no change in MSE. The gain is small but consistent, improving $62.6\%$ of the held-out cases with a signed-rank $p=2.2\times10^{-7}$, whereas over-weighting the structural term reverses it. Two ablations bound the effect. Adding any third model to the two-model ensemble degrades it, and classical unsharp masking fails to improve SSIM at any strength (best $0.8765$ against $0.8767$), so the gain reflects learned rather than indiscriminate sharpening. The result is a cheap, reproducible post-processing stage that improves an already strong ensemble without any large-scale retraining.
Read the full abstract
Analysis of PDEs
Decay Rates and Domain Dependence of a Coupled Wave-Heat System with Spatially Dependent Heat Coefficients
We study of the long-term behavior of a coupled wave-heat system. The system consists of a wave equation and a heat equation on two adjacent Lipschitz domains coupled by a common interface, with the heat equation being allowed to incorporate spatially dependent coefficients. We first establish strong asymptotic stability independent of the domains and coefficients. To this end, we employ the framework of closure relations, which reduces the spectral analysis of the coupled system to that of a wave equation. Secondly, we analyze non-uniform decay rates for classical solutions to the coupled system. Using a non-orthogonal decomposition of the state space, we reduce the problem to a resolvent-type estimate which is independent of the heat domain and heat coefficients. With this, we extend the known non-uniform decay rates to spatially dependent heat coefficients, yielding logarithmic decay under no assumptions and polynomial decay under the Geometric Control Condition.
Read the full abstract
General Relativity and Quantum Cosmology
Eight Local Couplings of Gravitational Waves from Unified Field Equations
A gravitational-wave (GW) detector records a linear mixture of local couplings under the assumption that extra polarizations enter geodesic deviation. Vacuum general relativity (GR) admits two transverse-traceless (TT) amplitudes. It remains to determine the largest set of couplings that can sit in that mixture, and whether a tensor-only arm-length result selects GR uniquely. The little group \(E(2)\) of a null four-momentum classifies the strain amplitudes \(\mathbf{p}=(p_{+},p_{\times},p_{x},p_{y},p_{b},p_{\ell})\), commonly written \(h_{P}\), which determine the electric tidal tensor along a ray. Geodesic deviation, recorded as differential arm length, therefore contains only those \(p_{P}\) that enter that tensor. Lorentz mixing at helicity \(\pm1\) supplies a gravito-magnetic (GEM) field that does not enter the tidal tensor. if GEM field is static, that does not propagate as a wave. A time-varying helicity-\(\pm1\) current sources a GEM wave that enters the mixture as a coupling, to be isolated by its measured quantity. For a radiation-zone wave that depends only on retarded time, \(\betag_{\perp}=\hat{\mathbf{k}}\times\partial_{t}(p_{x},p_{y})\), and the eight couplings are \(\Pvec=(\mathbf{p},\betag_{\perp})\). Here we adopt unified field equations on \(\mathbf{p}\) to clarify the origin of each component of \(\Pvec\); the measured quantity of each coupling then isolates the polarizations in that mixture, which favors identification of distinct polarizations and model tests.
Read the full abstract
Distributed Computing
Barnacle: Adaptive Multi-Leader Scheduling for DAG-Based Consensus
In DAG-based consensus, all validators propose blocks concurrently, and designated leader blocks drive transaction commit. Having multiple leader slots per round cuts queuing latency, yet production deployments run a single leader because of head-of-line blocking: a slow leader stalls the pipeline for at least one leader timeout, and for several waves when its slot must wait for the fallback indirect decision rule. This risk grows with the leader count. We introduce Barnacle, an add-on that adapts the leader count at run time. Every interval, it measures on the agreed committed DAG the fraction of slots decided as commit by the direct rule, and drives the leader count with additive increase, multiplicative decrease. The measurement requires no extra messages and no cryptography, and is deterministic. Barnacle is generic over DAG protocols; we instantiate it on four protocols spanning the Byzantine (3f + 1, 5f + 1), crash-only (2c + 1), and mixed (5f + 3c + 1) fault models, with proven safety and liveness. Results show Barnacle matches the best static leader count in every regime: in a healthy network its latency is 6-13% lower than a single leader's, and under degradation it matches a single leader while remaining 35-56% below a static high count. We are currently collaborating with the Sui team to integrate Barnacle into the Sui blockchain.
Read the full abstract
Signal Processing
Cooperative Multi-Task Semantic Communication for Joint Classification and Regression Tasks
Multi-Task semantic communication (SemCom) prioritizes simultaneous execution of multiple tasks over bit-accurate reconstruction in future intelligent networks. In our prior work [1], we introduced the cooperative multi-task SemCom (CMT-SemCom) framework, in which the semantic encoder is divided into a common unit (CU) and multiple specific units (SUs) to facilitate cooperative multi-task processing. However, CMT-SemCom has been evaluated on homogeneous classification tasks on simplistic datasets, limiting its applicability to real-world perception systems. In this paper, we extend our CMT-SemCom to jointly handle heterogeneous classification and regression tasks on the complex Cityscapes dataset. We adopt the information maximization (InfoMax) principle so that it accommodates mixed discrete and continuous semantic variables. In particular, we benchmark the proposed framework against independent single-task training, a conventional task-agnostic digital transmission, and single-encoder multi-decoder SemCom. Additionally, we investigate the impact of CU capacity on joint task performance, providing design insights. Extensive evaluations demonstrate that CMT-SemCom significantly outperforms the benchmarks.
Read the full abstract
Algebraic Geometry
A Polynomial PDE Criterion for the $-n/d$ Root of the Bernstein-Sato polynomial of Homogeneous Ideals
Let $I\subseteq\mathbb C[x_1,\ldots,x_n]$ be an ideal generated by homogeneous polynomials of a common degree $d$. We give a polynomial partial differential equation criterion guaranteeing that $-n/d$ is a root of the Bernstein-Sato polynomial $b_I(s)$. We apply this criterion to the ideal of maximal minors of a generic $m\times n$ matrix and obtain the distinguished root $-n$; combined with local divisibility along determinantal strata, this allows us to obtain the strong monodromy conjecture in the maximal-minor case. Finally, we prove that the criterion is stable under enlarging the linear span of the generators, adjoining generators in disjoint variables, products satisfying the natural slope condition, and Thom-Sebastiani sums. These stability results provide new classes of homogeneous ideals and polynomials for which the distinguished Bernstein-Sato root can be detected. Keywords. Bernstein-Sato polynomial, monodromy conjecture.
Read the full abstract
Earth and Planetary Astrophysics
Detection of hydrocarbons in Titan using high-resolution cross-correlation spectroscopy
High-resolution cross-correlation spectroscopy (HRCCS) is a powerful technique for detecting molecules whose individual spectral lines are too weak to be identified directly, but its sensitivity is limited by the availability and quality of high-resolution opacity data. Many molecules of atmospheric and astrobiological interest lack complete line lists, restricting traditional template-based searches. In this work, we use Titan as a controlled testbed to develop and validate a new cross-section-based methodology for HRCCS template construction. We analyse K-band CRIRES+ observations of Titan (1.99 - 2.48 {\mu}m), and compute cross-correlation functions using both line-by-line and cross-section-based templates. Our analysis recovers known hydrocarbons such as methane (CH4) and acetylene (C2H2), and yields the first HRCCS detection of ethane (C2H6) with a significance of SNRpeak = 5.17 {\pm} 0.07. The ethane detection was made possible exclusively through cross-section-based templates, as no high-resolution line list currently exists for this molecule. These results demonstrate that cross-section-based template construction is a practical and powerful strategy for extending HRCCS to molecules that currently lack reliable line lists, and establish Titan as a benchmark for calibrating molecular detection techniques that can be applied to both solar system and exoplanet atmospheres. Future applications of this approach to other ground-based high-resolution spectrographs, as well as to JWST's highest-resolution modes and next-generation facilities such as the ELT, could significantly expand the inventory of molecules detectable in planetary atmospheres.
Read the full abstract
Analysis of PDEs
An exact traveling waves solution for a special class of semilinear equations
We construct exact traveling-wave solutions for a semilinear parabolic equation with a specific nonlinearity. This nonlinearity arises, in particular, in voting models based on branching Brownian motion. Previously known results are special cases of those presented here.
Artificial Intelligence
Common-Witness Certificates and Sharp Feature Bounds for Counterfactual Image Auditing
An image editor may satisfy every regional plausibility constraint separately even when no single latent explanation fits the complete output. We formalize this local-to-global failure using a common witness grade and witness nerve. The framework separates auditing from causal identification: shared exogeneity alone allows every coupling of the regime marginals, whereas an externally justified witness relation yields sharp partial-identification bounds for prespecified image features. Helly-type arguments provide short incompatibility certificates for quasiconvex losses, heterogeneous action strata, and finite witness atlases; a blocker-hypergraph formula gives exact repair counts. Simultaneous confidence regions for the regime marginals give finite-sample outer coverage of the complete identified interval. Controlled MNIST, Morpho-MNIST, and smallNORB studies demonstrate the predicted local-global separation, while synthetic experiments test sharp bounds, certificate recovery, and structured computation. The method audits a declared feature relation and does not identify unrestricted pixel-level counterfactuals.
Read the full abstract
Machine Learning
OSR: Output Space Redistribution for Adaptive Label Removal in Classification Models
Label removal occurs frequently in classification systems with evolving taxonomies, where categories must be dynamically updated or eliminated. To accommodate such changes, classification models must adapt accordingly. Existing solutions, broadly categorized as retraining-based and feature-space-adjustment-based, share common limitations despite their variations, including reliance on access to original data, substantial computational and storage costs, inconsistent results, poor scalability, and degradation of model utility. To address this, we propose a novel approach that leverages statistical redistribution in the output space to approximate the post-removal confidence vectors of a retrained model. Applicable as a modular output filter, our method bypasses the burden of feature-space adjustments or loss-function convergence, alleviating scalability limitations. Furthermore, by requiring only existing labels and prior output confidences, the method potentially mitigates privacy concerns inherent to data-dependent solutions. Extensive experiments demonstrate competitive performance against full retraining, with improvements in computational efficiency and privacy preservation across several classification tasks.
Read the full abstract
Information Retrieval
The WebKurator.de Platform: Combined Regional and Topical Web Curation
The systematic curation of the Web remains a central challenge for national libraries and memory institutions that aim to preserve culturally and regionally relevant content. Existing directory-based approaches such as Curlie implement a predominantly topic-centric, one-dimensional hierarchy, where geographic aspects are intertwined with topical and linguistic categories. To address this limitation, we present WebKurator.de, a collaborative platform for combined regional and topical web curation, initially focused on the German web. WebKurator introduces a two-dimensional curation model that explicitly separates topical categorization and geographic annotation. The system integrates LLM-based topic classification and imprint-based address extraction with geocoding, and supports user suggestions together with moderated review. The platform is bootstrapped from the German Imprints Dataset, a large-scale collection of 5.54 million websites. Among them, 3.14 million contain imprint pages, for which we successfully extracted and geocoded postal addresses. Of these, 2.58 million (85.17%) are located in Germany and also have an assigned topic label. These websites form the initial foundation of WebKurator.de and can be continuously extended through user suggestions.
Read the full abstract
Robotics
Automated Weld Seam Recognition and 3D Mapping for Robotic Post Processing Using Photogrammetry and Semantic Segmentation
Accurate identification of weld seam geometries is essential for automated robotic post processing operations such as grinding, finishing, and inspection. For large workpieces, complete surface scanning using high precision laser scanners or structured light sensors can be time consuming and often generates substantial amount of data that are not relevant. This paper presents an experimental vision based pipeline for the approximate localization of weld seams. This serves as a preliminary stage before high precision measurement. The proposed approach aims to reduce the overall scanning effort and data acquisition efficiency. The proposed method includes capturing images of the workpiece from multiple viewpoints, identifying weld seams from the images using semantic segmentation, reconstructing the workpiece using photogrammetry, and projection of identified weld seams into the reconstructed model.
Read the full abstract
Fluid Dynamics
Rotational covariance, continuous spectrum and non-modal growth of three-dimensional planar weak viscous shock layers
Three-dimensional linear stability of a weak shock layer is analyzed with exact Reynolds-number-independent base-flow and stability equations, written in shock viscous scales. For any base flow depending on the shock-normal coordinate alone, with zero transverse velocity, the eigenvalue problem is proved covariant under rotation of the transverse wavenumber vector. Its eigenvalues depend only on the magnitude of this vector, and every eigenmode decouples exactly into an in-plane acoustic-entropy branch, which carries all dilatation and thermodynamic coupling, and an out-of-plane vortical branch, governed by a single shear-diffusion equation. The continuous spectra are obtained using the Grosch-Salwen far-field companion matrices and, independently, closed-form dispersion relations. The two agree to machine precision. The vortical continuous spectrum retreats from the real axis exactly as the square of the transverse wavenumber, while the acoustic-entropy continuous spectrum is strictly damped and, numerically, its least-damped point does not increase with this wavenumber. Three-dimensional disturbances are therefore damped at least as strongly as their two-dimensional counterparts with the same in-plane wavenumber, as measured by the least-damped point of the spectrum, exactly for the vortical branch and numerically otherwise, making the reduction the exact, Reynolds-free counterpart of Squire's transformation, in which the wavenumber magnitude plays the role of the Reynolds number. Planar oblique shock stability results are obtained from normal-shock results by a Doppler shift. Non-modal growth, measured in the Euclidean norm, is far stronger in the post-shock than in the pre-shock gas. Linear optimal disturbances of the shock-layer ripple the post-shock density field without displacing the shock front.
Read the full abstract
Probability
A one-parameter family of local-time conditioned processes : the generalized Brownian burglars
We construct and characterize the one-parameter family of real-valued processes that allow to recover the law of a Brownian loop-soup of any intensity on the real line conditionally on its occupation time field. These processes generalize the Brownian burglar constructed by Warren and Yor (that corresponds to our process in the limiting case where the intensity vanishes and there is just one Brownian motion). These processes are closely related to the Bass-Burdzy flow and recent work of Aïdékon, Hu and Shi on the stochastic Jacobi flow. Our approach uses a formalism, building on the notion of driver processes, where one considers the evolving object to be the line that we “stretch", instead of the burglar itself and its remaining occupation time. This leads to a characterization of these (generalized) Brownian burglars by simple natural axioms. It also allows to treat “negative intensities” and provides fairly direct derivations of several properties of the burglars, and in particular a “target-independence/locality" property at the special intensity where the local time is the square of a Gaussian Free Field.
Read the full abstract
Computation and Language
Investigating the Ability of Large Language Models to Analyze Recipes for Diabetes
Several studies have evaluated the ability of Large Language Models (LLMs) for meal planning, yielding positive outcomes. These models can process natural language inputs and leverage learned knowledge from their pretraining to generate meal plans. In this work, we investigate the ability of LLMs to analyze the suitability of given recipes for diabetes. The primary challenge for LLMs is to retrieve relevant dietary guidelines for diabetes, decompose recipes into ingredients and cooking methods, and apply these guidelines to determine the recipe's suitability. To study these challenges, we employ three kinds of prompts namely, (i) Direct Query Prompt (ii) Context-Guided Prompt, and (iii) Exemplary Context Prompt that incorporate different levels of diabetes dietary guidelines from medical sources. We introduce a benchmark dataset curated for this investigation consisting of 7607 recipes that include 3807 recipes suitable for diabetes and 3800 recipes not suitable for diabetes. Our results demonstrate that most LLMs are cautious in predicting recipes as suitable to prevent detrimental outcomes. Further, the models that can reason using the dietary guidelines performed better in predicting the suitability of recipes for diabetes. Overall, Mistral-7B and Llama 70B showed superior performance to their counterparts.
Read the full abstract
Artificial Intelligence
Interface-Induced Trajectory Censoring
Tool-use evaluations typically measure calls through a serving stack, which can report zero calls even when model outputs contain call-shaped payloads. We study interface-induced trajectory censoring: a mismatch between model emissions and the interface that converts them into actions. On selected BFCL v4 subsets, changing only the serving adapter while holding model weights and other evaluation settings fixed moves scores under the official scorer from 0.00 to 0.96 on simple_python and from 0.00 to 0.19 on multi_turn_base. A template-parser factorial experiment shows strong complementarity: neither component replacement alone restores parsing from the baseline, whereas their joint replacement recovers 196/200 first-request parses. Across five Qwen2.5-Coder checkpoints, classifier-positive emissions increase with model size while the mismatched interface reports no calls; a separate matched-interface Qwen3 ladder exhibits few silent emissions through 8B. Experiments on 115 tau-bench retail tasks extend the mechanism to a stateful, non-code environment. In a preregistered, single-seed, 150-step 7B LoRA experiment, changing only the parser registered in verl increases tool executions from 10 to 16,844. However, neither a shared programmatic-feedback evaluation nor a post-hoc common native function-calling evaluation detects a held-out success gain from repair; rescue counts do not increase. Interface repair thus restores tool-mediated experience without a detected performance gain in this setting. These findings show that tool-use measurements depend on the model-interface configuration and motivate layer-wise observability and conformance checks before attributing evaluation or training outcomes to the model alone.
Read the full abstract
Dynamical Systems
The finiteness conjecture for equilibria of electric fields generated by point charges of one sign
We prove that the electric field generated in three-dimensional space by finitely many point charges of one sign has only finitely many equilibrium points, thereby answering a 1969 question of Morse and Cairns (restated by Eremenko in 2008 and, as a conjecture, by Shapiro in 2015). More generally, for nonzero charges $q_i$ of possibly mixed signs at distinct sites $\mathbf a_i\in\mathbf{R}^3$, we show that the Coulomb field has at most $2^{N-4}(N-1)(9N^2+9N+10)$ equilibria in the region where $S(\mathbf x):=\sum_iq_i|\mathbf x-\mathbf a_i|^{-3}$ does not vanish. Of course, for charges of one sign, $S$ is nonzero everywhere. The proof rules out curves of equilibria using algebraic geometry and complex analysis on an associated complex curve, and then applies a Bézout count to obtain a quantitative bound. Well-known examples show that, in the mixed-sign case, the Coulomb field can vanish on curves contained in the zero set of $S$.
Read the full abstract
High Energy Astrophysical Phenomena
Ultra-compact twin stars with hybrid equations of state from bosonic dark matter
The properties of compact stars with a strong first-order phase transition to quark matter and with an additional fluid of self-interacting bosonic dark matter (DM) are studied. We find that the inclusion of DM changes the stability of mass-radius configurations relative to the naive one-fluid criterion. For compact star configurations with similar masses and different radii, so-called twin stars, the presence of DM removes the unstable segment between the hadronic and the hybrid branch, so that the stable mass-radius sequence becomes continuous after the onset of the phase transition to quark matter. We furthermore find stable ultra-compact objects (UCOs), defined by a total compactness $C = M_\text{tot}/R_\text{grav} \ge 1/3$. We observe two distinct classes of UCOs: a DM-halo class with $f_\text{DM} \gtrsim 0.9$, and a DM-core class at $f_\text{DM} \lesssim 0.02$. The two classes can be separated by the surface redshift of the normal matter, which reaches $z = 0.734 - 0.808$ for the DM-core and extreme values of $z = 2 - 3.468$ for the DM-halo class, allowing us to use their redshift as a way to disentangle the two classes. Furthermore, the high redshift of the DM-halo UCOs could serve as a signal of DM since their redshift exceeds the Buchdahl limit of $z = 2$. Finally, we find hybrid star solutions of “ultimate twins" with similar masses and visible radii, but different dark matter content, leading to different tidal deformabilities and surface redshifts. Future X-ray and gravitational measurements of ultra-compact neutron stars with radii and masses outside the allowed neutron star range can thereby probe the presence and the properties of DM in addition to a first-order phase transition to quark matter.
Read the full abstract
Group Theory
On the surjectivity conjectures of Dupont and Monod
In this paper we apply our work on polylogarithmic cocycles representing the Borel classes to establish that the comparison map from bounded continuous cohomology to continuous cohomology is surjective for connected semisimple Lie groups with finite centre. This proves conjectures of Dupont and Monod.
Read the full abstract
Analysis of PDEs
$L^p$ isomorphisms for the fractional Laplacian on $\mathbb{R}^n$ and their application
A classical theory of Amrouche, Girault and Giroire (1994) resolves the Laplace equation on $\mathbb{R}^n$ through an isomorphism between weighted Sobolev spaces. Inspired by their framework, we develop the corresponding $L^p$ theory for the fractional Laplacian: we introduce weighted fractional Sobolev spaces $\Lambda^{s,p}(\mathbb{R}^n)$, gauged at top order by $\|(-\triangle)^{\frac{s}{2}}u\|_{L^p}$, and prove that $(-\triangle)^{\frac{s}{2}}:\Lambda^{s,p}(\mathbb{R}^n)/\mathcal P_{[s-n/p]}\to L^p(\mathbb{R}^n)$ is an isomorphism for all $s\in(0,2)$ and $p\in(1,\infty)$; thus $(-\triangle)^{\frac{s}{2}}u=f$ is solvable for every $f\in L^p(\mathbb{R}^n)$, uniquely modulo an explicit finite-dimensional space of polynomials. More generally, extending the scale by duality to negative orders, the fractional Laplacians of the appropriate orders map its spaces isomorphically onto one another and compose exactly. A key ingredient is a family of weighted Hardy and Poincaré inequalities adapted to $\|(-\triangle)^{\frac{s}{2}}u\|_{L^p}$, established here for $1\le s<2$ by a reduction to the gradient. Since these inequalities hold a priori only on a dense class, while the density of $C_{\mathrm{c}}^\infty(\mathbb{R}^n)$ in $\Lambda^{s,p}(\mathbb{R}^n)$ is itself nontrivial, we first prove the isomorphism on the closure of the Schwartz class, which coincides with $\Lambda^{s,p}(\mathbb{R}^n)$ since the difference of two solutions is an entire $s$-harmonic function, hence a polynomial; the density follows as a by-product. As an application, we obtain existence and uniqueness, with explicit kernels and compatibility conditions, for the fractional Stokes system in $\mathbb{R}^n$.
Read the full abstract
Number Theory
On the exponential sum over squarefree integers
Let $\mu$ be the Möbius function and $e(t)=e^{2\pi it}$. We prove that if $N\ge2$, $\alpha\in\mathbb{R}$, $(a,q)=1$, and $|\alpha-a/q|\le q^{-2}$, then \[\bigg|\sum_{n\le N}\mu^2(n)e(\alpha n)\bigg|\ll\left(\frac Nq+q\right)(\log 2N)^5, \] with an absolute implied constant, and we deduce the corresponding estimate on the minor arcs of the Hardy–Littlewood dissection throughout the range $Q\le N^{1/2}$. The estimates of Schlage-Puchta [SP] and of Tolev [T] have the same dependence on $q$ and $Q$ but carry a factor $N^{\varepsilon}$. The proof uses Heath-Brown's square sieve with sieving primes confined to an interval $(P,2P]$, where $P$ may be as small as a multiple of $\log N$; a finite Fejér majorant in place of a truncated Fourier series; and, after completion of the character sums, a count of representations that exploits the restriction on the primes in place of the divisor function.