2026-10论文中文摘要
你何时可以剪枝你的网络？多语言语音解析中中间神经元的研究
When Can You Prune Your Network? A Study of Intermediate Neurons in Multilingual Speech Parsing
arXiv 2610.11520 · 2026-10-09残差优势：用于可验证奖励强化学习的学生相对教师指导
Residual Advantage: Student-Relative Teacher Guidance for RL with Verifiable Rewards
arXiv 2610.11519 · 2026-10-09体-边界分解与精确理想流体恒星分布
Bulk-boundary decompositions and exact perfect fluid stellar distributions
arXiv 2610.11518 · 2026-10-09Bermond–Thomassen 猜想的一个渐近界
An asymptotic bound for the Bermond--Thomassen Conjecture
arXiv 2610.11517 · 2026-10-09带次模目标的拟阵秘书问题的精细分析
A Refined Analysis for Matroid Secretary with Submodular Objectives
arXiv 2610.11516 · 2026-10-09基于神经网络与参数辨识的交通及行人模型的随机梯度下降算法
Stochastic Gradient Descent for Traffic and Pedestrian Models via Neural Networks and Parameter Identification
arXiv 2610.11515 · 2026-10-09SSCBench：评估使用工具的大语言模型智能体的故障注入测试的证据有效性
SSCBench: Evaluating the Evidential Validity of Fault-Injection Tests for Tool-Using LLM Agents
arXiv 2610.11514 · 2026-10-09隐藏路径的验证：量子密钥分发网络的可扩展拓扑保证
Certifying Hidden Paths: Scalable Topology Assurance for QKD Networks
arXiv 2610.11513 · 2026-10-09用于细胞极化的体-表面反应扩散系统中脉冲解的存在性
Existence of pulse solutions in a bulk--surface reaction--diffusion system for cell polarization
arXiv 2610.11512 · 2026-10-09EIFL：在联邦学习中针对不可信服务器高效保护全局模型的隐私与完整性
EIFL: Efficiently Protecting Global Model Privacy and Integrity Against an Untrusted Server in Federated Learning
arXiv 2610.11511 · 2026-10-09现代标准阿拉伯语（MSA）在大语言模型（LLMs）中是否主导阿拉伯方言？一项表示层面的分析
Does Modern Standard Arabic (MSA) Dominate Arabic Dialects in LLMs? A Representation-Level Analysis
arXiv 2610.11510 · 2026-10-09带Neumann型边界条件的一阶非线性发展方程的凸性保持
Convexity preservation for first-order nonlinear evolution equations with Neumann-type boundary conditions
arXiv 2610.11509 · 2026-10-09WARP-VLA：面向视觉-语言-动作模型中视角鲁棒策略执行的腕部相机适配
WARP-VLA: Wrist-Camera Adaptation for View-Robust Policy Execution in Vision-Language-Action Models
arXiv 2610.11508 · 2026-10-09基于跨半径一致性学习的单声道连续半径区域语音提取
Monaural Continuous-Radius Regional Speech Extraction with Cross-Radius Consistency Learning
arXiv 2610.11507 · 2026-10-09紧致性与一致性：深度图聚类的联合框架
Compactness and Consistency: A Conjoint Framework for Deep Graph Clustering
arXiv 2610.11506 · 2026-10-09DAMP：基于去噪信念学习与对抗运动先验的人形机器人运动控制
DAMP: Humanoid Locomotion via Denoised Belief Learning and Adversarial Motion Priors
arXiv 2610.11505 · 2026-10-09算子不匹配问题：基于便携式GPU计算部署BEV感知
The Operator Mismatch Problem: Deploying BEV Perception with Portable GPU Compute
arXiv 2610.11504 · 2026-10-09氯取代对CrSBr磁光性质的影响
Effects of chlorine substitution on the magneto-optical properties of CrSBr
arXiv 2610.11503 · 2026-10-09Fed-GRPO：奖励信号驱动的联邦组相对策略优化
Fed-GRPO: Reward-Signal-Driven Federated Group Relative Policy Optimization
arXiv 2610.11502 · 2026-10-09超越序列：为LLM推荐提炼结构化决策记忆
Beyond Sequences: Distilling Structured Decision Memory for LLM Recommendation
arXiv 2610.11501 · 2026-10-09在介质傅里叶空间中跨超表面家族学习qBIC共振
Learning qBIC Resonances across Metasurface Families in Dielectric Fourier Space
arXiv 2610.11500 · 2026-10-09dtμ分子离子近阈共振对有效原子2s-2p劈裂的响应
Response of the near-threshold resonances of the $\bm{dtμ}$ molecular ion to an effective atomic $2s$--$2p$ splitting
arXiv 2610.11499 · 2026-10-09用于少步视频生成的参数化轨迹蒸馏
Parametric Trajectory Distillation for Few-Step Video Generation
arXiv 2610.11498 · 2026-10-09通用半单重构与障碍理论
Universal Semisimple Reconstruction and Obstruction Theory
arXiv 2610.11497 · 2026-10-09停止我的舞蹈！理解、检测与归因运动感知深度伪造视频
Stop My Dancing! Understanding, Detecting and Attributing Motion-Aware Deepfake Videos
arXiv 2610.11496 · 2026-10-09腔介导自旋链中的拓扑超辐射普适类
Topological Superradiant Universality Classes in Cavity-Mediated Spin Chains
arXiv 2610.11495 · 2026-10-09液体中的激光碎裂：构建通用反应图谱
Laser fragmentation in liquid - constructing a generic reaction map
arXiv 2610.11494 · 2026-10-09来自不稳定D-膜的渐近平直时空量子引力
Quantum Gravity in Asymptotically Flat Spacetimes from Unstable D-Branes
arXiv 2610.11493 · 2026-10-09面向能量采集物联网的能量感知接收端发起式MAC协议
Energy-Aware Receiver-initiated MAC Protocol for Energy Harvesting IoT
arXiv 2610.11492 · 2026-10-09关于误差界与非临界拉格朗日乘子
On error bounds and noncritical Lagrange multipliers
arXiv 2610.11491 · 2026-10-09用于协同自动驾驶中高效后量子消息认证的零知识签名框架
A Zero-Knowledge Signature Framework for Efficient Post-Quantum Message Authentication in Cooperative Automated Driving
arXiv 2610.11490 · 2026-10-09超越分辨率：实例检索中的对象-图像比例不匹配
Beyond Resolution: Object-to-Image Ratio Mismatch in Instance Retrieval
arXiv 2610.11489 · 2026-10-09MARC：针对编解码器攻击的自回归音频生成多比特水印
MARC: Multi-Bit Watermarking for Autoregressive Audio Generation against Codec Attacks
arXiv 2610.11488 · 2026-10-09短区间内Möbius函数与线性相位的不相关性
Discorrelation of the Möbius function with linear phases in short intervals
arXiv 2610.11487 · 2026-10-09PSI-SINDy：非线性动力学稀疏识别的后选择推断
PSI-SINDy: Post-Selection Inference for Sparse Identification of Nonlinear Dynamics
arXiv 2610.11486 · 2026-10-09利用基本矩阵特征值的欧氏球面中紧致极小子流形的最佳挤压定理 II
An optimal pinching theorem on compact minimal submanifolds in the Euclidean spheres via eigenvalues of fundamental matrices II
arXiv 2610.11485 · 2026-10-09从能带重构到玻戈留波夫色散：dz2能带如何增强铁基超导电性
From band reconstruction to Bogoliubov dispersion: How dz2-band enhances iron-based superconductivity
arXiv 2610.11484 · 2026-10-09BridgeGuard：基于扩散模型的自动驾驶的显式安全漂移
BridgeGuard: Explicit Safety Drift for Diffusion-based Autonomous Driving
arXiv 2610.11483 · 2026-10-09评估用于可复现数据工程的本地语言模型智能体：一项关于移动性工作流的实证软件工程研究
Evaluating Local Language Model Agents for Reproducible Data Engineering: An Empirical Software Engineering Study of Mobility Workflows
arXiv 2610.11482 · 2026-10-09伯努利作业的随机调度问题的准多项式时间近似方案
A QPTAS for Stochastic Scheduling of Bernoulli Jobs
arXiv 2610.11481 · 2026-10-09RoboAware：基于反事实结果学习具身技能的协调方法
RoboAware: Learning to Coordinate Embodied Skills from Counterfactual Outcomes
arXiv 2610.11480 · 2026-10-09条件残差预测：无需双向教师模型改进自回归视频扩散模型
Conditional Residual Prediction: Improving Autoregressive Video Diffusion without a Bidirectional Teacher
arXiv 2610.11479 · 2026-10-09驯服空间的加倍、庞加莱与格罗莫夫-豪斯多夫预紧性
Doubling, Poincaré and Gromov-Hausdorff precompactness for tamed spaces
arXiv 2610.11478 · 2026-10-09大规模超导量子处理器的并行校准与抗漂移维护的定量规则
Quantitative Rules for Parallel Calibration and Drift-Resilient Maintenance of Large-Scale Superconducting Quantum Processors
arXiv 2610.11477 · 2026-10-09软X射线光束线自动控制终端站中数据的生成、积累与利用
Generation, Accumulation, and Utilization of Data in the Automatically Controlled End-Station of a Soft X-ray Beamline
arXiv 2610.11476 · 2026-10-09罕见门分歧会限制可塑性：梯度流何时会错误预测有限批次SGD
Rare Gate Disagreements Can Limit Plasticity: When Gradient Flow Mispredicts Finite-Batch SGD
arXiv 2610.11475 · 2026-10-09近壁集中翼型格栅实现湍流管流的净减阻
Net drag reduction in turbulent pipe flow by airfoil-section grids concentrated near the wall
arXiv 2610.11474 · 2026-10-09基于ETH的对偶化问题的拟多项式下界
An ETH-based quasipolynomial lower bound for Dualization
arXiv 2610.11473 · 2026-10-09从思维链到循环：基于循环Transformer的非自回归潜在推理
From Chain-of-Thought to Loops: Non-Autoregressive Latent Reasoning via Looped Transformers
arXiv 2610.11472 · 2026-10-09操作系统安全研究综述
A Survey of Security Research for Operating Systems
arXiv 2610.11471 · 2026-10-09
