High Energy Astrophysical Phenomena
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Monday 14 September
Fast radio bursts (FRBs) are millisecond-duration extragalactic pulses whose discovery requires searches over large signal-parameter spaces and the rejection of candidate sets dominated by radio-frequency interference and noise. Integrating quantum processors into FRB searches requires a workflow connecting telescop data to hardware-compatible models. Here we develop a pipeline that converts raw search-mode PSRFITS data into dynamic-spectrum segments and identifies FRB-like signals using a hybrid Quantum Vision Transformer (QViT). On labelled FAST data, the selected QViT achieves a mean accuracy of 94.00% and recall of 98.30%, with similar performance to a compact classical Vision Transformer. Recall reaches 97.11% on an FRB source excluded from training and model selection. Real-device execution on 82 segments yields a mean simulator-hardware Hellinger fidelity of 0.9161 +/- 0.0074. The workflow connects raw telescope data to quantum-assisted identification and provides an experimental basis for assessing the measurement requirements that must be addressed before survey-scale deployment.
General Relativity and Quantum Cosmology
Interior gravitational perturbations to naked singularities of a scalar field
For the $k$-self-similar naked singularity solutions of Einstein–scalar field equations constructed by Christodoulou, we construct a family of interior gravitational perturbations leading to trapped surface formation that remain large but finite at the threshold Hölder regularity, while converging to zero in all regularities below the threshold. This extends the interior spherically symmetric result below the threshold in \cite{Li25} to non-spherically symmetric setting. We further construct a genuinely localized family of perturbations whose angular support shrinks to a single point, with smooth convergence to zero away from that point. This exploits the additional angular freedom available beyond spherical symmetry. Moreover, when measured instead in Sobolev regularity, such genuinely localized family of perturbations still converges to zero in all regularities below the same threshold.
Read the full abstract
Computation and Language
When the Wrong Key Wins: Understanding and Detecting Hallucinations in LLMs
Large language models can hallucinate even when the knowledge required for a correct answer is already available. We study this failure through a latent-key view of inference, where answer selection depends on competition among associations acquired during pretraining. We show that model predictions can be highly sensitive to individual query keywords, that these influential keywords exhibit entity-specific binding, and that their effects are systematically shaped by pretraining frequency. Multiple bindings can also compete and exhibit higher-order interactions within the same query. Based on this mechanism, we introduce a two-stage keyword-perturbation method for hallucination detection. By removing influential keywords and measuring how the model reorganizes its prediction, the method distinguishes errors caused by misleading key associations from correct decisions supported by diagnostic evidence. Across multiple models and benchmarks, perturbation provides a strong and transferable detection signal, reaching $0.910$ AUROC on probe-known ScientistQA. Finally, we extend the same probabilistic framework to four hallucination regimes: knowledge deficit, wrong knowledge, context distraction, and unstable inference. Their operational distributions across benchmarks provide diagnostic context for why different detector families succeed in different settings.
Read the full abstract
Materials Science
Re4+ Luminescence as a Highly Sensitive Alternative to Ruby for Optical Pressure Sensing
Luminescence-based remote pressure sensing provides a powerful approach for pressure determination under conditions where conventional contact methods are difficult to implement. Although ruby remains the undisputed gold standard among luminescent pressure indicators, its relatively low pressure sensitivity and susceptibility to temperature variations represent important limitations. Therefore, in this work, Cs2HfCl6:Re4+ is proposed as an alternative luminescent pressure indicator exhibiting a 15-fold higher pressure sensitivity (6.93 nm GPa-1) than ruby while maintaining a comparable thermal sensitivity, making it a particularly attractive material for optical manometry. Furthermore, Cs2HfCl6:Re4+ enables ratiometric pressure readout, providing a high relative pressure sensitivity reaching 175.1% GPa-1. To the best of our knowledge, this work represents the first demonstration of Re4+ luminescence for pressure sensing, introducing a new class of luminescent pressure indicators and opening new opportunities for the development of highly sensitive optical manometers based on Re4+-activated materials.
Read the full abstract
Fluid Dynamics
Physics Informed Neural Network model for the dynamical study of Abdominal Aortic Aneurysm
We present the development and application of a three-dimensional Physics-Informed Neural Network (PINN) framework for the investigation of haemodynamic behaviour in the human aorta. The model incorporates a time-resolved simulation of pulsatile blood flow over a two-minute interval, enabling the extraction of pressure and velocity fields with high temporal fidelity. The mechanical stress exerted on the aortic wall was quantified through Laplace's law, with temporal averaging applied to derive representative stress distributions. This approach circumvents the computational overhead associated with conventional computational fluid dynamics (CFD) methods by eliminating mesh generation and exploiting the automatic differentiation capabilities inherent to neural networks. The proposed methodology demonstrates that PINNs can serve as an efficient and accurate alternative for modelling complex vascular flow phenomena, offering significant advantages in scalability and computational cost reduction while maintaining physical consistency.
Read the full abstract
Analysis of PDEs
Rigidity for two-phase overdetermined problems in $\mathbb{S}^n$
In this paper, we study rigidity phenomena associated with a class of overdetermined two-phase problems on the $n$-dimensional sphere $\mathbb{S}^n$. Specifically, we analyze the solutions of coupled elliptic equations defined on a domain $\Omega \subset \mathbb{S}^n$ and its complement, subject to Dirichlet boundary conditions on the interface $\partial\Omega$ and a compatibility condition on the gradient. We utilize two different methods. The method of moving planes works when $\partial\Omega$ is contained inside an open hemisphere. The other method works when $\Omega$ is simply connected in $\mathbb{S}^2$. Both lead to the same conclusion. Namely, the existence of a solution to such overdetermined problems necessarily implies that $\Omega$ is a geodesic ball. In the latter, we prove this rigidity property for three distinct scenarios: a baseline problem with piecewise constant source terms, which is the two-phase version of the torsion problem; a generalization including a class of positive and regular nonlinearities; and finally an eigenvalue problem involving the first Dirichlet eigenvalue of the respective phases.
Read the full abstract
Materials Science
Multiple Linear Regression-Enhanced RGB-based Luminescence Thermometry for High-sensitivity Thermal Readout
The practical implementation of phosphors for luminescence-based thermal sensing and imaging requires simple, user-friendly approaches that enable convenient and sensitive temperature readout. To the best of our knowledge, this is the first demonstration of combining RGB-based thermal imaging using a conventional digital camera with multiple linear regression (MLR) to achieve straightforward and highly sensitive temperature determination and spatially resolved thermal imaging. Importantly, the implementation of the MLR approach enhances the relative sensitivity by more than 3-fold, compared with conventional analysis based solely on the G/R or B/R intensity ratios. The application of Ca3Al2O6:Mn2+-Ce3+ as a luminescent temperature probe, in which the intensity ratio between the Ce3+ and Mn2+ emission bands exhibits a pronounced temperature dependence, enables temperature readout through several complementary approaches. These include conventional luminescence intensity ratio thermometry (SR = 1.46% K-1), analysis of the CIE 1931 chromaticity coordinates (SRx = 0.45% K-1 and SRy = 0.14% K-1), as well as RGB-based thermal sensing and imaging using a digital camera. This multimodal optical response, together with the accessibility of camera-based readout and the substantial sensitivity enhancement enabled by MLR, establishes a practical strategy for spatially resolved luminescence thermometry.
Read the full abstract
Accelerator Physics
High-Intense Gamma-Ray Emission from a Crystalline Undulator with Realistic Bending Profiles
The development of compact and intense $\gamma$-ray sources in the MeV energy range remains a significant frontier in radiation physics, with profound implications for nuclear physics, medicine and applied science. In this work, we present a comprehensive numerical investigation of the photon emission probability and brilliance of a Crystalline Undulator (CU) based on a periodically bent Si(110) crystal. Our approach integrates realistic deformation profiles obtained via Finite Element Method simulations of a realistic sample, where bending is induced by patterned Si$_3$N$_4$ surface stressors. Relativistic molecular dynamics simulations, performed using the MBN Explorer software package, consider a 10 GeV positron beam consistent with the foreseen FACET-II facility specifications. Our results reveal a distinct undulator radiation peak in the 1.6-2.1 MeV range, well-separated from the broader channeling radiation background. We show that for an optimal aperture angle of $1/2\gamma$, the source reaches a maximum peak brilliance of about $5\times10^{22}$~photons/s/mm$^2$/mrad$^2$/0.1%~BW. This performance is highly competitive with large-scale Gamma-Beam Systems and exceeds that of Inverse-Compton Scattering sources, confirming the potential of crystalline undulators as high-brilliance, compact light sources for the MeV domain.
Read the full abstract
Computer Vision
ChatGPT Images 2.5 in the Wild: A Launch-Period Dataset and Detector Evaluation
An image tool can change its underlying generator while retaining its public name, making version attribution from online posts ambiguous. We study this problem after the ChatGPT Images 2.5 launch. Our frozen collection contains 3,478 images from 2,440 posts across 8 sources. Recorded posting times fall within the first 51.1 hours after the announcement. It records three attribution tiers and retains standalone images after image-form filtering and targeted review. Caption claims and host records provide admission evidence, not independently verified generator identity. The observed content profile depends on the source mixture: NightCafe supplies 39.0% of images but 77.0% of CLIP-assigned fantasy scenes. We then evaluate six frozen detectors at thresholds calibrated to a 5% flag rate on reference photographs. Collection flag rates range from 3.7 to 56.4%, falling 42-81 percentage points below GenImage recall. Held-out artwork false-positive rates range from 1.5 to 96.5%, so a higher collection flag rate does not by itself establish better detection. An exploratory X-only comparison with our April collection finds a higher September flag rate for Effort, and a suggestive difference for DoU, under fixed-threshold post-clustered bootstrap intervals. Attribution, content and processing differences prevent a causal interpretation of these contrasts. The collection supports analysis of reported model use during a product transition, with source and attribution evidence retained for interpretation. The collection is released at https://scam.ai/research.
Read the full abstract
Quantum Physics
Biorthogonal Time-Dependent Variational Principle for Non-Hermitian Systems
We develop a biorthogonal time-dependent variational principle for real-time dynamics of non-Hermitian quantum many-body systems. Independent left and right matrix-product states obey coupled bivariational tangent-space equations whose cross-Gram matrix defines an oblique projection. A matrix-free scaled Taylor action propagates the resulting non-normal local generators without assembling dense matrices or storing a Krylov basis. We distinguish the fully coupled algorithm, which solves the cross-pairing problem and truncates the two bond bases jointly, from an efficient independently propagated approximation used for large systems. Independent truncation can make the retained left-right pairing nearly singular; overlap drift and the smallest singular value of the bond cross matrix expose this failure, while coupled truncation substantially delays it. Exact benchmarks and convergence tests validate the method. Applied to an interacting long-range non-Hermitian Ising chain, it resolves a biorthogonal dynamical quantum phase transition and shows that a weak imaginary field shifts the leading critical time from $t^{\ast}|J|=1.84$ to $1.04$.
Read the full abstract
Computer Vision
LG-VLN: A Zero-Shot Vision-and-Language Navigation Framework with LangGraph State Orchestration
Continuous-environment vision-and-language navigation (VLN-CE) requires interpreting natural-language instructions in unseen 3D environments and executing continuous low-level actions. Existing methods often depend on LiDAR, panoramic cameras, or extra sensors; separate geometric-mapping and semantic-navigation visual representations can cause long-trajectory spatial-semantic inconsistencies. We propose LG-VLN, a monocular zero-shot framework with shared visual features and LangGraph-based state orchestration. An online feed-forward 3D reconstruction network predicts depth, camera poses, and dense point clouds for agent-pose estimation and global map fusion. Geometry and navigation share dense CleanDIFT features: semantic consistency rejects incorrect inter-frame correspondences, while target-instance constraints define visual references whose similarity combines with local BLIP-2 image-text relevance to form a semantic value map. LangGraph represents instruction parsing, geometric perception, semantic value updates, path planning, action execution, and failure recovery as a directed state graph with conditional transitions, persistent state, and modular recovery mechanisms. On a fixed 550-episode subset of the R2R-CE val-unseen split, LG-VLN achieves 21.3% success and 12.1% success weighted by path length. Ablations show shared semantic features improve navigation, further boosted by combining visual similarity and image-text relevance. Results establish shared visual representations and explicit state orchestration as effective for zero-shot VLN-CE using monocular RGB alone. Code will be publicly released for reproducibility.
Read the full abstract
Cryptography and Security
CounterPersona: Append-Only Defense Against Unauthorized Persona Skill Distillation
Persona skill distillation can extract recurring patterns from personal information and encode them into reusable skills, enabling AI systems to closely replicate an individual's behavior. However, such replication also raises serious concerns regarding personal privacy and labor autonomy. Unlike existing perturbation-based defenses that require individuals to modify their data before collection, once historical records are collected by an attacker, they can no longer be altered, sanitized, or revoked. Therefore, such defenses are difficult to adapt to this append-only setting. To solve this challenge, we introduce CounterPersona, which constructs targeted counter-persona evidence, packs compatible behavioral states into compact realization units, and strengthens them through rationale-guided consistency rewriting. We conduct extensive experiments showing that CounterPersona achieves strong and consistent effectiveness across lexical, semantic, and LLM-based measures, while remaining robust across distillers. Our work establishes a skill anti-distillation paradigm for protecting personal privacy and labor autonomy against unauthorized skill distillation.
Read the full abstract
Artificial Intelligence
OpenAI4S: Code as Action, Science as Sessions
AI co-scientists could accelerate computational research, but over a long-running study the workflow also has to stay inspectable, resumable and reproducible, which requires persistent computational state and provenance. Here we present OpenAI4S, an open-source scientific research agent built around the principle of Code as Action, Science as Sessions. OpenAI4S combines a persistent computing runtime with research-session management: orchestration is handled through structured tool calls, while scientific actions are represented as complete code cells executed in persistent Python and R kernels. An append-only Action Ledger, per-cell execution records, versioned artifacts, environment records, and workspace checkpoints preserve how results were produced and support session recovery, branching, and extension. Configurable sandboxing, permission controls, and code and trajectory screening provide complementary safeguards. We evaluate OpenAI4S on 36 research scenarios spanning retrosynthesis, molecular dynamics, protein binder design, protein mutation, catalyst screening, and mineral spectroscopy, measuring scientific task accuracy, workflow completeness, and reproducibility of the resulting repositories. OpenAI4S achieves an overall score of 7.83, compared with 5.7–6.4 for a general-purpose coding harness evaluated with three frontier models, with the largest gains on long-horizon and computation-intensive workflows. These results suggest that integrating persistent execution with session-level provenance can improve the reliability of AI-assisted scientific workflows. Environment specification and full rerunnability remain weak for every evaluated system, ours included, so reproducibility is still an open problem for scientific agents. The system is available under the MIT license at \href{https://github.com/PKU-YuanGroup/OpenAI4S}{github.com/PKU-YuanGroup/OpenAI4S}.
Read the full abstract
Analysis of PDEs
Optimal Time-Decay of global solutions to the Navier-Stokes-Maxwell system
We prove the existence and uniqueness of global-in-time solutions to the Navier–Stokes–Maxwell (NSM) system for small initial data in the critical Fujita–Kato space $\dot H^{\frac{d}{2}-1}(\mathbb R^d)$, in any dimension $d\geq 3$. Under the additional assumption that the initial data belong to the Besov space $\dot B^{-\frac{d}{2}}_{2,\infty}(\mathbb R^d)$, we also establish the optimal decay rate $t^{-\frac{s}{2}-\frac{d}{4}}$ at infinity of the solution in $\dot H^s(\mathbb R^d)$, for any $s\in\left(-\frac{d}{2},\frac{d}{2}-1\right]$. This is achieved by constructing a Lyapunov functional that is equivalent to the pointwise energy on the Fourier side, and by combining it with an adaptation of the Fourier splitting method to establish its optimal decay rate. All the analysis carried out here - from the global existence theory to the study of the large-time behavior - is performed within a framework that is uniform with respect to the speed of light $c\in(0,\infty)$. In particular, in the non-relativistic limit $c\to\infty$, this allows us to recover the same results for the corresponding limiting magnetohydrodynamic system.
Read the full abstract
Information Retrieval
Generate to Explore, Select to Exploit: Aligning LLM-based Headline Generation with Personalized Recommendation
In industrial recommendation feeds, presenting a static headline for an item often fails to satisfy the diverse, multimodal interests of the user population, particularly suppressing the needs of long-tail audiences. While Large Language Models (LLMs) have been integrated into recommendation for content understanding or ranking, directly optimizing them to output a single best headline typically leads to mode collapse—converging to generic patterns that satisfy average tastes but miss specific latent intents. To bridge this gap, we introduce GESE (Generate to Explore, Select to Exploit), a framework operating at the system's presentation layer that decouples personalization into generative exploration and selective exploitation. First, we treat the LLM as a probabilistic explorer, utilizing Group Sequence Policy Optimization (GSPO) with a hierarchical reward mechanism to generate a candidate set that maximizes the semantic coverage of potential user interests. Subsequently, a lightweight, real-time feedback-aware selector acts as the exploiter, identifying the optimal realization from the candidate pool based on instant contextual signals. Extensive deployment on a commercial platform with over 100 million daily active users demonstrates that GESE significantly outperforms state-of-the-art baselines, achieving a 2.57% lift in CTR and 0.87% in dwell time. These results validate that decoupling diversity-oriented generation from precision-oriented selection offers a robust blueprint for aligning generative AI with dynamic user utility.
Read the full abstract
Soft Condensed Matter
Spontaneous Vortex Instability in Active Nematics
One of the defining results in the study of active matter is the spontaneous flow instability, through which a homogeneous, uniformly aligned state breaks translational symmetry along a single direction and develops sustained flow. The vortex state that emerges at higher activity has instead been attributed to nonlinear dynamics. Using a Floquet-type linear stability analysis, we show that no such mechanism is required: the flowing state undergoes a secondary, zigzag instability that breaks the remaining translational symmetry and produces the vortex state. We further identify a regime in which the flowing state ceases to exist and vortices emerge directly from the uniformly aligned state. Under channel confinement, the instability selects a length scale that differs from the establishedactivelengthscale, andsetsthenumberofvorticesthatappear, leadingtoaconfinement- selected pattern reminiscent of a vortex lattice, opening a route toward direct experimental tests of this instability. Full nonlinear simulations reproduce the predicted onset activities and the selected vortex number.
Read the full abstract
Disordered Systems and Neural Networks
Bias and Correlations in Quasiperiodicity: Impact on Localization in an Extended Aubry-André Model
Unlike in Anderson localization - where any amount of uncorrelated disorder localizes all eigenstates in one dimension - a one-dimensional system with quasiperiodic potential supports a richer range of localization behavior. This paper investigates the fundamental question of which potential characteristics govern localization properties. We characterize quasiperiodic potentials using two independent parameters, correlation and bias, and demonstrate that bias, in addition to correlation, critically influences localization. Through the study of an extended Aubry-André model, in which a tunable parameter allows systematic control over both the bias and correlation of the potential, we show that bias is key in determining the fraction of delocalized states. An increase in the potential strength generally enhances the tendency toward localization, while simultaneously strengthening the correlations in the quasiperiodic potential. This apparent counterintuitive behavior can be understood in terms of the bias parameter: increasing the potential strength reduces the bias, which in turn favors localization. For the family of Hamiltonians considered, we identify two critical bias thresholds: below the lower threshold, the entire spectrum is localized, whereas below the higher threshold, at most a fraction of the states can be delocalized, precluding delocalization of the entire spectrum. To further test our framework, we examine a quasiperiodic model with zero bias and find that, despite its correlated nature, all states become localized even at very weak potential strengths - recovering the Anderson-like scenario.
Read the full abstract
Pattern Formation and Solitons
Deformation of sine-Gordon two-soliton solutions in $\varphi^4$ kink-antikink configurations
In this work, we construct analytical kink-antikink $(K\bar{K})$ configurations in the non-integrable $\varphi^4$ model by mapping exact solutions of the integrable sine-Gordon system via a field deformation. This procedure yields two distinct classes of configurations, parametrized by the initial half-separation and velocity. We compare these profiles with the standard additive superposition Ansatz in terms of vacuum structure and equation-of-motion residuals, deriving explicit closed-form expressions for the corresponding integrated squared residuals. For small separations, the mapped soliton-antisoliton configurations exhibit appreciably smaller residuals than the naive superposition Ansatz, which in turn performs better than the mapped two-soliton configurations. Although the mapped fields are not exact solutions of the $\varphi^4$ equation of motion, they provide mathematically consistent, topologically sound, and physically motivated initial data for numerical studies of kink-antikink scattering and resonance phenomena.
Read the full abstract
Dynamical Systems
Classification of Stationary Configurations of Four Identical Point Vortices
We find and classify every stationary configuration of the system of four identical point vortices in the plane. Up to the natural symmetries of the system, there are exactly three configurations, which are relative equilibria: a square, a collinear configuration, and an equilateral triangle with a central vortex. Among them, the square configuration is the global minimum of the Hamiltonian, whereas the other two configurations are saddle points. In particular, the square is the unique stable configuration among the stationary configurations. Using the special algebraic structure of four points, we introduce a change of variables based on the Hadamard matrix of order four, which transforms the original four-vortex problem into an equivalent three-point problem.
Read the full abstract
Artificial Intelligence
ER-EDF: A Psychology-Grounded Emotion Regulation Framework for Speech Empathetic Dialogue Generation in Large Audio-Language Models
In human conversation, empathy involves not only recognizing a partner's emotion but also regulating it, listeners identify the emotional state, select a suitable regulation strategy, and implement it in their reply. Current large audio-language models (LALMs) largely map perceived emotion directly onto a response, skipping this regulation process. As a result, their responses tend to mirror the user's affect or offer generic sympathy. We propose ER-EDF, a psychology-grounded framework for empathetic dialogue that follows the three stages of Gross's Extended Process Model of emotion regulation, identifying the user's emotion from speech, selecting a regulation strategy matched to that emotion, and implementing the strategy through a LALM. The framework is model-agnostic and integrates seamlessly into existing LALMs. We further construct a spoken empathetic dialogue dataset and introduce empathy-aware evaluation metrics beyond lexical matching. Experiments across five LALMs and two datasets show that ER-EDF consistently improves empathetic response quality in both automatic and human evaluations. These results suggest that modeling how humans regulate others' emotions, rather than only how they perceive them, is key to empathetic spoken dialogue, paving the way for psychologically grounded empathetic AI.
Read the full abstract
Complex Variables
Asymptotic expansions of $L^2$-extension indices and curvature positivity
In this paper, we prove an asymptotic expansion of the $L^2$-extension index of a smooth Hermitian metric on a holomorphic vector bundle, in which the Chern curvature appears as the second-order coefficient. By using this expansion, we show in a unified way that there is an equivalence between how sharp the $L^2$-extension is and how positive or negative the curvature is. We also introduce a new notion of $q$-$L^2$-extension indices and investigate partial positivity and flatness in terms of these indices.
Read the full abstract
Machine Learning
Beyond Numerical Time Series: A Unified Benchmark for Multimodal Forecasting with Heterogeneous Context
Most time series forecasting benchmarks remain numerical-centric and provide limited support for evaluating contextual information that shapes real-world temporal dynamics. Existing multimodal benchmarks also suffer from limited data and context coverage, fragmented evaluation settings, and overreliance on aggregate evaluation. In this paper, we propose MUSE-Bench, a unified benchmark for multimodal time series forecasting with heterogeneous context. It comprises fourteen datasets across eight domains and six types of context: metadata, events, holidays, news, images, and numerical covariates. We evaluate diverse forecasting paradigms, including statistical, data-specific, foundation, multimodal, and general-purpose LLM forecasting methods under shared non-overlapping forecast windows, common target observations, and consistent point and probabilistic metrics. Extensive experiments yield three main findings. First, numerical time series foundation models dominate the overall ranking, while Aurora, the evaluated multimodal foundation model, trails the leading numerical TSFMs but outperforms all evaluated data-specific models. Second, ablations show that external context improves the four evaluated context-aware models, whereas incorrect or temporally misaligned context degrades performance. Third, general-purpose LLMs perform poorly as direct forecasters, and LLM-guided refinement does not yield consistent improvements. MUSE-Bench enables systematic evaluation of how forecasting models utilize context and provides a foundation for future multimodal forecasting research.
Read the full abstract
Combinatorics
Thresholds and Fluctuations for Colorful Arithmetic Progressions in Sparse Random Colorings
In this paper, we derive thresholds and fluctuations for arithmetic progressions with prescribed color patterns in sparse random colorings of $[n]:=\{1, 2, \ldots, n\}$, where each element of $[n]$ is colored independently according to a given probability vector. For any admissible ordered palette of colors, we determine the full multi-parameter threshold region for the appearance of a colorful arithmetic progression. The threshold is governed by two competing mechanisms: a global first-moment condition and a local color-availability condition, resulting in a polyhedral satisfiability region, with a piecewise-polyhedral threshold surface. In the satisfiability region we establish asymptotic normality for the number of colorful arithmetic progressions of a given length, with an explicit rate of convergence in Wasserstein distance. On the threshold surface, we identify three distinct asymptotic regimes: Poisson, compound Poisson with mixed Poisson jumps, and compound Poisson with uniform jumps, after an appropriate normalization. These results provide a complete description of the threshold and fluctuation behavior of general colored arithmetic progressions under sparse random colorings, in a unified framework that interpolates between classical uncolored/monochromatic progressions in binomial random subsets and multicolored, including rainbow, arithmetic progressions.
Read the full abstract
Game Theory
Improved Impossibility Bounds for Maximin Share Allocations
The maximin share (MMS) is a central fairness benchmark for allocating indivisible items, but it need not be simultaneously attainable even under additive preferences. While extensive work has developed approximation guarantees, quantitative impossibility bounds have received comparatively little attention. We establish improved asymptotic and constant impossibility bounds for both goods and chores. For every sufficiently large number $n$ of agents, we construct additive goods instances in which every allocation gives some agent at most a $1-\Omega((\log n)^{-2})$ fraction of her MMS. This strengthens the $1/n^4$ shortfall of Feige, Sapir, and Tauber (2021) to an inverse-polylogarithmic shortfall. For chores, we construct instances in which every allocation gives some agent cost at least a $1+\Omega((\log n)^{-2})$ factor of her MMS. Consequently, for every fixed $\varepsilon>0$, guarantees of $1-O(n^{-\varepsilon})$ for goods and $1+O(n^{-\varepsilon})$ for chores are impossible. Both asymptotic constructions use $n+O(\log n)$ items. Together with concurrent existence results of Feige (2026), this determines the threshold $n+\Theta(\log n)$ for guaranteed MMS existence under additive preferences, for both goods and chores. We also give four-agent, eleven-item instances that improve the universal impossibility bounds from $39/40$ to $20/21$ for goods and from $44/43$ to $31/30$ for chores.
Read the full abstract
Numerical Analysis
Shape calculus and automatic differentiation for multi-phase level-set topology optimisation with unfitted finite elements
We present shape calculus techniques and a scalable automatic shape differentiation framework for multi-phase topology optimisation on unfitted discretisations defined by several level-set functions. First we establish general, exact shape calculus expressions in the discrete case for multi-phase systems by leveraging concepts from convex geometry. To complement this theoretical foundation, we introduce an open-source multi-phase automatic shape differentiation framework based on polytopal cutting. This computational framework is validated against both finite differences and our established exact expressions, matching the latter to near machine precision. Furthermore, the proposed automatic shape differentiation is scalable across distributed computing environments, demonstrating near ideal weak scaling up to 1.65 billion finite elements across 13,824 computer cores. We demonstrate our implementation by solving unfitted multi-phase topology optimisation problems for anisotropic diffusion, linear elasticity, and fluid–structure interaction. Together, these theoretical and computational contributions provide a robust and accessible foundation for advancing multi-phase topology optimisation using unfitted finite element methods. In particular, the methods enable the solution of topology optimisation problems involving multi-phase and multi-physics systems with non-trivial boundary conditions. The open-source software is available at https://github.com/zjwegert/GridapTopOpt.jl.
Read the full abstract
Machine Learning
$\mathbb{SL}(n)$ Representation Learning: An Intrinsic Mixed-Curvature Space with Higher Curvature Capacities and Deeper Order-Aware Composition
Mixed-curvature representation learning seeks to capture rich geometric structures that cannot be adequately modeled by a single curvature regime. Existing approaches largely rely on product manifolds, which require manually specifying how different curvature spaces are combined and separate their curvature contributions across factors. We introduce the $\mathbb{SL}(n)$ space, a representation geometry defined by the simple $\det(A)=1$ constraint and a left invariant Schatten-$p$ Finsler structure. Despite this minimal construction, $\mathbb{SL}(n)$ exhibits pointwise negative, zero, and positive flag curvature around a common flagpole, while its mixed-curvature and curvature-coupling capacities are asymptotically maximal relative to the intrinsic geometric upper bound. Beyond geometry, its noncommutative group structure provides inherent order sensitivity, and its non-nilpotent Lie algebra admits nonzero nested Lie brackets at arbitrary depth, enabling deep order-aware composition. Empirically, $\mathbb{SL}(n)$ consistently outperforms a broad range of representation manifold baselines across graph benchmarks at different scales. It reduces average distortion over the strongest baselines by $44.3\%$ on KEGG and $40.5\%$ on HumanCyc, and improves Hits@20 by $42.8\%$ on OGBL-PPA. Experiments on Flickr30k-Order further support its ability to capture higher order dependencies from ordered composition. Together, these results show how a seemingly simple structural constraint can yield unexpectedly rich geometry, capacity, and composition within a unified representation space.
Read the full abstract
Robotics
Task-Distribution-Aware Counterweight Synthesis and Constrained Co-Design for Serial Manipulators
Passive counterweights are simple gravity compensators, but a counterweight selected from a single pose is not generally optimal for the configurations and tasks a manipulator actually executes. This paper develops a task-distribution-aware synthesis framework in which the operating distribution $\rho(q)$ enters the design explicitly. For a counterweight moment $p=m_c r_c$ with gravity torque $-gp\phi(q)$, the weighted mean-square residual gravity torque has the closed-form minimizer $p^*=E_\rho[\tau_g\phi]/(gE_\rho[\phi^2])$. If payload gravity torque is affine in payload mass, the optimum is also affine: $p^*(m_p,\rho)=p_0^*(\rho)+m_pK_p(\rho)$. For fixed static moment, added counterweight inertia is $I_c=pr_c$ while mass is $m_c=p/r_c$, so mass-radius selection is underdetermined unless physical constraints are specified. A recovered three-link manipulator is used as a case study. At $r_c=0.20$ m, zero-payload equivalent optima are 0.672 kg for uniform joint-space operation, 0.683 kg for approximately uniform task-space operation, 0.713 kg for a representative pick-and-place family, and 0.952 kg for a high-gravity-biased distribution, a change of more than 40% caused solely by the operating distribution. Nondominated fronts show that preferred mass-radius pairs depend on declared engineering bounds. A rated-torque-referenced all-joint screen increases zero-payload feasible task-space coverage from 78.1% without compensation to 93.7% for the uniform-distribution design. A lumped point-mass trajectory study gives a provisional crossover from no counterweight at very aggressive motion to stronger compensation as motion slows. These actuator and dynamic results are engineering consequence studies rather than physical validation.
Read the full abstract
Combinatorics
Obstructions to coloring arithmetic graphs
The arithmetic graph $B_n$ joins distinct $a,b\in\N$ when $\max(a,b)/\gcd(a,b)\le n$. We prove $\chi(B_{205})=206$, disproving the conjecture that $\chi(B_n)=n$ for every $n$, equivalently the Rainbow Cascades Conjecture. The proof reduces an arbitrary tiling by the arithmetic exponent tile to a periodic tiling, then to two families of finite quotients, which are excluded using exact computations. We also construct a $208$-coloring using $\Z_{104}\times\Z_2$ and prove $212\le\chi(B_{211})\le213$. The lower bound at $211$ follows from prime-cardinality tiling rigidity and the published nonexistence of a cyclic logarithm of length $211$; we give a direct proof of the required rigidity statement. Finally, we record the equivalence with the List Cascade Coloring Conjecture and the conjecture on ironic decorations, and deduce finite graph counterexamples to both. The least $n$ with $\chi(B_n)>n$ is either $195$ or $205$; determining which remains open.
Read the full abstract
Rings and Algebras
The Köthe conjecture via point modules
We give a short construction refuting the Köthe Conjecture. The construction is based on the point module approach to the problem, introduced by the first-named and second-named authors together with Easton and Ford.
Computation and Language
Translating the Translator: Decomposing the Cost of English-Forced Inter-Agent Communication
Multi-agent LLM architectures, such as LangChain and AutoGen, largely assume English as the lingua franca for internal inter-agent communication, even when the end-user task is non-English. We fill this gap by evaluating a two-agent extraction-answer core, with an additional back-translation agent in the English-forced condition, across four typologically diverse languages (Hindi, Chinese, Spanish, Arabic; n = 300 per language) using the Aya-23-8B model. We compare a native-language pipeline to an English-forced one (which incorporates a final back-translation step from English to the user's language). We discover a statistically significant English-Forcing Tax (surviving a strict Bonferroni correction) that isolates the cost of English routing from general multi-agent orchestration overhead. Forcing inter-agent communication through English reduces Exact Match accuracy by 13.0 percentage points (Spanish) up to 30.6 percentage points (Hindi) compared to native-language multi-agent execution. Using chrF scores as a diagnostic measure of English-reference lexical overlap, we find that lower overlap is strongly associated with pipeline failure, consistent with translation loss being an important contributor to the observed performance drop. These findings suggest a compelling case for native-language routing in agent frameworks when the source and target languages are typologically distant, reducing a compounding translation tax.