2026-10论文中文摘要
陷阱门 Clifford 算子及其应用
Trapdoored Clifford Operators and Applications
arXiv 2610.01848 · 2026-10-02使用LLM作为验证器推理检测模型规范中的不一致性
Detecting Inconsistencies in Model Specifications with LLM-as-Verifier Reasoning
arXiv 2610.01847 · 2026-10-02超越可解码性：声学因素是否驱动基于语音的阿尔茨海默病评估中的预测？
Beyond Decodability: Do Acoustic Factors Drive Predictions in Speech-Based Alzheimer's Assessment?
arXiv 2610.01846 · 2026-10-02龙棋的时间差分学习
Temporal-Difference Learning for Dragonchess
arXiv 2610.01845 · 2026-10-02齐次开放量子随机游走的弹道速度与势论常返性
Ballistic Speed and Potential-Theoretic Recurrence for Homogeneous Open Quantum Random Walks
arXiv 2610.01844 · 2026-10-02最优随机双层优化与一阶预言机
Optimal Stochastic Bilevel Optimization with First-Order Oracles
arXiv 2610.01843 · 2026-10-02时间序列世界模型中准确性与机制一致性的分歧
On the Divergence of Accuracy and Mechanism Consistency in Time Series World Models
arXiv 2610.01842 · 2026-10-02ILW和BO方程的量子层级及曲线模空间上的DR环
Quantum hierarchies of the ILW and BO equations and the DR cycles on the moduli spaces of curves
arXiv 2610.01841 · 2026-10-02RFBniCS：用于液流电池的开源仿真框架
RFBniCS: An open-source simulation framework for redox flow batteries
arXiv 2610.01840 · 2026-10-02非线性多项式无散向量场的导出代数
The Derived Algebra of Nonlinear Polynomial Divergence-Free Vector Fields
arXiv 2610.01838 · 2026-10-02Artin特征的平均值及其与Davenport--Heilbronn定理的关系
Averages of Artin characters and their relation to the Davenport--Heilbronn theorem
arXiv 2610.01837 · 2026-10-02联合DESI DR1全形状功率谱与双谱分析中的宇宙学推断
Cosmological inference from a joint DESI DR1 full-shape power spectrum and bispectrum analysis
arXiv 2610.01836 · 2026-10-02Varda-single-1.0：瑞士复杂地形上空1公里分辨率下的确定性数据驱动天气预报
Varda-single-1.0: deterministic data-driven weather forecasting at 1 km resolution over Switzerland's complex topography
arXiv 2610.01835 · 2026-10-02代码掌控模拟，Jev 掌控评估
Code Owns the Simulation, Jev Owns the Evaluation
arXiv 2610.01834 · 2026-10-02持续过程级评估：面向演进中的企业AI智能体技能
Continuous Process-Level Evaluation for Evolving Enterprise AI Agent Skills
arXiv 2610.01833 · 2026-10-02惠勒延迟选择实验的双重选择实验实现
Experimental realization of Wheeler's delayed-choice experiment with dual selections
arXiv 2610.01832 · 2026-10-02池化有帮助，学习权重在上下文学习中反而有害：分解组注意力
Pooling Helps, Learned Weighting Hurts In-Context: Decomposing Group Attention
arXiv 2610.01831 · 2026-10-02通量-和隧穿宇称控制的约瑟夫森非互易性：并联双量子点中单重态-三重态竞争
Flux- and tunnel-parity-controlled Josephson nonreciprocity from singlet-triplet competition in a parallel double quantum dot
arXiv 2610.01830 · 2026-10-02精细紧化万有雅可比簇 $\overline{J}_2$ 和 $\overline{J}_{2,1}$ 的上同调
Cohomology of the fine compactified universal Jacobians $\overline{J}_2$ and $\overline{J}_{2,1}$
arXiv 2610.01829 · 2026-10-02语言模型对齐的渐近行为与记忆效应
The Asymptotics of Language Model Alignment with Memory
arXiv 2610.01828 · 2026-10-02验证拥堵下的科学发现：基于多保真度成对排序的方法
Scientific Discovery under Validation Congestion via Multi-Fidelity Pairwise Rankings
arXiv 2610.01827 · 2026-10-02令牌通信辅助的协作具身人工智能：概念、框架与机遇
Token Communication-Assisted Collaborative Embodied Artificial Intelligence: Concepts, Framework, and Opportunities
arXiv 2610.01826 · 2026-10-02近阈值卡鲁扎-克莱因引力子衰变与暗维度
Near Threshold Kaluza-Klein Graviton Decays and the Dark Dimension
arXiv 2610.01825 · 2026-10-02QPI-DeepONet-MAC：一种可扩展且稳定的物理信息深度算子网络混合经典-量子架构
QPI-DeepONet-MAC: A Scalable and Stable Hybrid Classical-Quantum Architecture for Physics-Informed Deep Operator Networks
arXiv 2610.01824 · 2026-10-02广义Engression模型
Generalized Engression Models
arXiv 2610.01823 · 2026-10-02可微混合动作神经反馈控制用于区域供热网络
Differentiable Hybrid-Action Neural Feedback Control for District Heating Networks
arXiv 2610.01822 · 2026-10-02超越线性概念：发现并对齐大型语言模型中的非线性概念流形
Beyond Linear Concepts: Discovering and Aligning Non-Linear Concept Manifolds in Large Language Models
arXiv 2610.01821 · 2026-10-02有限数据下动态非对称驱动安全信息性
Finite-Data Safety Informativity Under Dynamic Asymmetric Actuation
arXiv 2610.01820 · 2026-10-02MECHVAR：方差引导的机制判别用于自主机器学习实验选择
MECHVAR: Variance-Guided Mechanism Discrimination for Autonomous Machine Learning Experiment Selection
arXiv 2610.01819 · 2026-10-02面向预测的后验分布的马尔可夫链蒙特卡洛方法
Markov chain Monte Carlo for predictively oriented posteriors
arXiv 2610.01818 · 2026-10-02NIR-EKF：基于归一化创新比率的扩展卡尔曼滤波器用于鲁棒状态估计
NIR-EKF: Normalized Innovation Ratio-Based EKF for Robust State Estimation
arXiv 2610.01817 · 2026-10-02系统光谱观测揭示$z\sim3$处多样性的原星系团
Diverse Protoclusters at $z\sim3$ Disclosed by Systematic Spectroscopy
arXiv 2610.01816 · 2026-10-02去偏一切：扩散模型中无需监督的公平性与多样性
Debias Anything: Fairness with Diversity without Supervision in Diffusion Models
arXiv 2610.01815 · 2026-10-02碰撞模型：马尔可夫与非马尔可夫杂质模型
Collision models: Markovian and Non-Markovian impurity models
arXiv 2610.01814 · 2026-10-02AI辅助有丝分裂计数提高多种肿瘤类型的可重复性和效率
AI-assisted mitotic counting improves reproducibility and efficiency across multiple tumour types
arXiv 2610.01813 · 2026-10-02识别符号式凸性是困难的
Recognizing Signomial Convexity is Hard
arXiv 2610.01812 · 2026-10-02星系原星系团作为宇宙再电离的驱动者：II. Lyman-α发射体的Te基金属丰度
Galaxy Protoclusters as Drivers of Cosmic Reionization: II. Te-Based Metallicities of Lyman-α Emitters
arXiv 2610.01811 · 2026-10-02连续时间Galton-Watson树爆炸附近的行为
Behaviour near explosion in continuous-time Galton-Watson trees
arXiv 2610.01810 · 2026-10-02自适应多分辨率扩散算子：演化多分辨率空间上的变分理论
Adaptive Multiresolution Diffusion Operators: A Variational Theory on Evolving Multiresolution Spaces
arXiv 2610.01809 · 2026-10-02通过半轻子衰变探测中微子质量
Probing the neutrino mass through semileptonic decays
arXiv 2610.01808 · 2026-10-02PhaseAT：用于医学图像域泛化的傅里叶相位对抗训练
PhaseAT: Fourier Phase Adversarial Training for Medical Image Domain Generalization
arXiv 2610.01807 · 2026-10-02闭测地线均为简单闭测地线的复双曲曲面
Complex hyperbolic surfaces whose closed geodesics are all simple
arXiv 2610.01806 · 2026-10-02机制设计中的鲁棒性
Robustness in Mechanism Design
arXiv 2610.01805 · 2026-10-02关于非交换多项式的秩零点猜想
On a Rank Nullstellensatz Conjecture for Noncommutative Polynomials
arXiv 2610.01804 · 2026-10-02圆β-系综迹的高斯极限
Gaussian limits for traces of circular beta-ensembles
arXiv 2610.01803 · 2026-10-02Baum-Connes猜想：一门简明课程
The Baum-Connes conjecture: a concise course
arXiv 2610.01802 · 2026-10-02安全原型并非安全方向：响应安全嵌入中的参考依赖与提示混杂
A Safe Prototype Is Not a Safety Direction: Reference Dependence and Prompt Confounds in Response-Safety Embeddings
arXiv 2610.01801 · 2026-10-02LineupRL：通过描述到序列识别实现时间序列描述的可验证强化学习
LineupRL: Verifiable Reinforcement Learning for Time Series Captioning via Caption-to-Series Identification
arXiv 2610.01800 · 2026-10-02权重准则、Dong 性质与 Operad 的 Manin 积
The Weight Criterion, Dong Property, and Manin Products of Operads
arXiv 2610.01798 · 2026-10-02旋转平衡中的广义涨落-耗散定理与爱因斯坦关系
Generalized fluctuation-dissipation theorem and Einstein relation in rotating equilibrium
arXiv 2610.01797 · 2026-10-02
