2026-10论文中文摘要
高斯消去法的不稳定性呈指数级罕见（部分结果的证明）
Instability of Gaussian elimination is exponentially rare (proof of partial result)
arXiv 2610.04761 · 2026-10-06碳化硅上外延石墨烯的位点特异性轨道对称性破缺
Site-Specific Orbital Symmetry Breaking in Epitaxial Graphene on Silicon Carbide
arXiv 2610.04760 · 2026-10-06基于时间重构约束的段路由流量工程
Segment Routing Traffic Engineering with Time-Based Reconfiguration Constraints
arXiv 2610.04759 · 2026-10-06EasyClassifier：为不编程的研究者提供的诚实、可复现的机器学习分类工具
EasyClassifier: Honest, Reproducible Machine-Learning Classification for Researchers Who Do Not Program
arXiv 2610.04758 · 2026-10-06提示一致性推理用于零样本流匹配文本到语音模型
Prompt-Consistency Inference for Zero-Shot Flow-Matching Text-to-Speech Models
arXiv 2610.04757 · 2026-10-06自进化大语言模型系统何时应停止进化？
When Is Enough Enough in Self-Evolving LLM Systems?
arXiv 2610.04756 · 2026-10-06混合整数有界约束二次分式问题的全局优化
Global Optimization of Mixed-Integer Bound-Constrained Quadratic Fractional Problems
arXiv 2610.04755 · 2026-10-06FoSeRL：强化学习策略的正式序列鲁棒性认证
FoSeRL: Formal Sequential Robustness Certification for Reinforcement Learning Policies
arXiv 2610.04754 · 2026-10-06方差感知的在线强化学习细粒度间隙依赖界
Variance-Aware Fine-Grained Gap-Dependent Bounds for Online Reinforcement Learning
arXiv 2610.04752 · 2026-10-06动态路由：大语言模型测试时多功能性的新维度
Dynamic Routing as a New Dimension for Test-time Versatility of LLMs
arXiv 2610.04751 · 2026-10-06群作用与旋量表示导致定义域大小为3的Holant*复杂性二分定理
Group Action and Spin Representation Lead to a Holant* Complexity Dichotomy on Domain Size 3
arXiv 2610.04750 · 2026-10-06从蒸馏私人健康记录中智能体发现血液生物标志物
Agentic discovery of blood biomarker from distilled private health records
arXiv 2610.04749 · 2026-10-06存在异常值时的对数预测密度分数估计
Log predictive density score estimation in the presence of outliers
arXiv 2610.04748 · 2026-10-06形式化言语行为的道德评价：真实性、谎言与伦理困境
Formalizing the Moral Evaluation of Speech Acts: Truthfulness, Lies and Ethical Dilemmas
arXiv 2610.04747 · 2026-10-06表格强化学习的精确快速批量模拟
Exact Fast Batch Simulation for Tabular Reinforcement Learning
arXiv 2610.04746 · 2026-10-06$\times 2 \times 3$ 本征函数结构中的熵率
Entropic rates from eigenfunction structure for $\times 2 \times 3$
arXiv 2610.04745 · 2026-10-06少说谎，说大谎：针对LLM机器人团队的隐蔽内部攻击
Lie Rarely, Lie Big: Stealthy Insider Attacks on LLM Robot Teams
arXiv 2610.04744 · 2026-10-06ThyCLIPNet：一种BiomedCLIP引导的轻量级注意力增强DeepLabV3+框架，用于稳健的甲状腺结节分割
ThyCLIPNet: A BiomedCLIP-Guided Lightweight Attention-Enhanced DeepLabV3+ Framework for Robust Thyroid Nodule Segmentation
arXiv 2610.04743 · 2026-10-06机器人学习与视觉预测力
Robot Learning with Visual Predicted Force
arXiv 2610.04741 · 2026-10-06迈向本地可部署的智能体协同科学家：面向早期药物发现的小模型规划
Toward a Locally Deployable Agentic Co-Scientist: Small-Model Planning for Early-Stage Drug Discovery
arXiv 2610.04740 · 2026-10-06关于具有弱Ulrich或初始化aCM切丛的极化曲面的分类
On the classification of polarized surfaces with weakly Ulrich or initialized aCM tangent bundles
arXiv 2610.04739 · 2026-10-06固定靶 $^{208}$Pb+$^{40}$Ar 超外围碰撞中激发的 $^{40}$Ar 核发射粒子（LHC）
Particle emission from $^{40}$Ar nuclei excited in fixed-target $^{208}$Pb+$^{40}$Ar UPC at the LHC
arXiv 2610.04738 · 2026-10-06PyINE：一种通过代码执行实现可扩展启发与监督的框架
PyINE: A Framework for Scalable Elicitation and Oversight via Code Execution
arXiv 2610.04737 · 2026-10-06手持手机的行人概率预测：世界坐标系热图、视觉惯性高度漂移及无真值评估
Probabilistic Pedestrian Forecasts from a Handheld Phone: World-Frame Heat Maps, Visual-Inertial Height Drift, and Evaluation without Ground Truth
arXiv 2610.04736 · 2026-10-06估算数据中心用水量：最佳实践与关键问题
Estimating data center water use: Best practices and critical questions
arXiv 2610.04735 · 2026-10-06通信网络中的拓扑无关量子优势
Topology-independent quantum advantage in communication networks
arXiv 2610.04734 · 2026-10-06基于高阶平均场控制屏障函数的领导者-跟随者密度安全控制
Safe Leader-Follower Density Control via Higher-Order Mean-Field Control Barrier Functions
arXiv 2610.04733 · 2026-10-06论病毒的传播与行波，第一部分
On the Spread of a Virus and Traveling Waves, Part I
arXiv 2610.04732 · 2026-10-06面向 NISQ 与容错硬件的矩阵模型动力学平方和编译
Sum-of-squares compilation of matrix-model dynamics for NISQ and fault-tolerant hardware
arXiv 2610.04731 · 2026-10-06扭曲仿射格拉斯曼流形中 Mirković-Vilonen 环的组合学
Combinatorics of Mirković-Vilonen cycles in twisted affine Grassmannians
arXiv 2610.04730 · 2026-10-06马约拉纳双线性量、紧致模种子与Tsirelson界的逼近
Majorana bilinears, compact modular seeds and the approach to the Tsirelson bound
arXiv 2610.04729 · 2026-10-06基于深度强化学习的自适应混合光射频传输用于超可靠低延迟通信
Adaptive Hybrid Optical RF Transmission Based on Deep Reinforcement Learning for Ultra-Reliable Low Latency Communication
arXiv 2610.04728 · 2026-10-06探究视频Transformer中的时空冗余性以实现碰撞预判
Investigating Spatiotemporal Redundancy in Video Transformer for Collision Anticipation
arXiv 2610.04727 · 2026-10-06通过最小化齐次分配代价实现全驱动多旋翼中的涌现欠驱动行为
Emergent Underactuation in Fully Actuated Multirotors by Minimizing a Homogeneous Allocation Cost
arXiv 2610.04726 · 2026-10-06Verb-ICL：重新思考结构化预测中的上下文学习
Verb-ICL: Rethinking In-Context Learning for Structured Prediction
arXiv 2610.04725 · 2026-10-06二维自由晶格费米海中各向异性弗里德尔振荡与等周输运
Anisotropic Friedel oscillations and isoperimetric transport in two-dimensional free lattice Fermi seas
arXiv 2610.04724 · 2026-10-06潜变量拉格朗日神经网络用于非自治非线性动力系统的降阶建模
Latent-Lagrangian Neural Networks for Reduced Order Modeling of Non-autonomous Nonlinear Dynamical Systems
arXiv 2610.04723 · 2026-10-06NAMVIS：下一尺度自回归多视图图像合成
NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis
arXiv 2610.04722 · 2026-10-06Knossos 与 Ariadne：使用视觉语言模型对完整图表拓扑提取进行基准测试与学习
Knossos and Ariadne: Benchmarking and Learning Complete Diagram Topology Extraction with Vision-Language Models
arXiv 2610.04721 · 2026-10-06WNet：用于高效令牌混合的离散小波变换
WNet: Discrete Wavelets Transform for Efficient Token Mixing
arXiv 2610.04720 · 2026-10-06学习在有限交互下澄清不完整意图
Learning to Clarify Underspecified Intents Under Limited Interaction
arXiv 2610.04719 · 2026-10-06学习OTA：边缘感知、计算与通信的统一框架
Learning OTA: A Unified Framework for Edge Sensing, Computation, and Communication
arXiv 2610.04718 · 2026-10-06太阳能电池中通过垂直无序钙钛矿纳米线实现高效光捕获策略
Efficient Light Trapping Strategy in Solar Cells via Vertically Disordered Perovskite Nanowires
arXiv 2610.04717 · 2026-10-06使用编码代理自动剪枝日志代码：我们进展如何？
Towards Automatically Pruning Logging Code with Coding Agents: How Far Are We?
arXiv 2610.04716 · 2026-10-06GPU加速的Bregman Douglas-Rachford分裂算法用于离散最优传输
GPU-Accelerated Bregman Douglas-Rachford Splitting for Discrete Optimal Transport
arXiv 2610.04715 · 2026-10-06波长依赖的光致发光映射研究苯乙铵四氟硼酸盐处理钙钛矿太阳能电池在热应力下的深度依赖降解
Wavelength-Dependent Photoluminescence Mapping of Depth-Dependent Degradation Under Thermal Stress in Phenethylammonium Tetrafluoroborate Treated Perovskite Solar Cells
arXiv 2610.04714 · 2026-10-06二维TEM重建磁性纳米颗粒形貌的隐藏代价
The Hidden Cost of Two-Dimensional TEM Reconstruction of Magnetic Nanoparticle Morphology
arXiv 2610.04713 · 2026-10-06Box-iLQR：通过有限障碍轨迹优化实现约束感知反馈
Box-iLQR: Constraint-Aware Feedback via Finite-Barrier Trajectory Optimization
arXiv 2610.04712 · 2026-10-06在多变量网络荟萃分析中识别跨多种治疗获益-危害特征的折中方案
Identifying compromise solutions across multiple treatment benefit-harm profiles in multivariate network meta-analysis
arXiv 2610.04711 · 2026-10-06组织语言模型智能体的轨迹证据：片段、方法与残余
Organising Trajectory Evidence for Language-Model Agent Assurance: Fragments, Methods, and the Residual
arXiv 2610.04710 · 2026-10-06
