2026-10论文中文摘要
当AI发现隐藏消息时，它会报告吗？
When AI Finds Hidden Messages, Does It Report?
arXiv 2610.10620 · 2026-10-09TestJack：你是否应该信任编码基准中的结果？通过评估器演化进行智能体编码基准审计
TestJack: Should you trust the results in coding benchmarks? Agentic Coding Benchmarks Auditing via Evaluator Evolution
arXiv 2610.10619 · 2026-10-09反极布尔函数的黑林格猜想
The Hellinger Conjecture for Antipodal Boolean Functions
arXiv 2610.10618 · 2026-10-09MRCert：通过类型特定掩码实现对抗性修补样本的部署后修补鲁棒性认证
MRCert: Towards Post-deployment Patch Robustness Certification for Adversarially Patched Samples via Type-specific Masking
arXiv 2610.10617 · 2026-10-09当路由暴露成员身份：来自MoE路由器遥测的隐私泄露
When Routing Reveals Membership: Privacy Leakage from MoE Router Telemetry
arXiv 2610.10616 · 2026-10-09JevForest：面向预算受限特征获取的路径投票方法
JevForest: Path Voting for Budgeted Feature Acquisition
arXiv 2610.10615 · 2026-10-09大整数参数下广义拉盖尔多项式的不可约性
Irreducibility of Generalised Laguerre Polynomials for Large Integer Parameters
arXiv 2610.10614 · 2026-10-09用于异常检测的带联邦轻量头的时间Transformer CAN编码器
Temporal transformer CAN encoder with federated lightweight heads for anomaly detection
arXiv 2610.10613 · 2026-10-09PyCache陷阱：智能体技能扫描器中的检查-执行间隙
PyCache Trap: The Inspection-Execution Gap in Agent Skill Scanners
arXiv 2610.10612 · 2026-10-09智能体AI中的验证与自我改进：基础与局限
Verification and Self-Improvement in Agentic AI: Foundations and Limits
arXiv 2610.10611 · 2026-10-09代码理解是编码智能体的瓶颈
Code Understanding is a Bottleneck for Coding Agents
arXiv 2610.10610 · 2026-10-09含三个变量的黎曼函数方程的一个推广
A Generalization of the Riemann Functional Equation with Three Variables
arXiv 2610.10609 · 2026-10-09从调查失败到可靠的安全运营中心智能体：理解和改进基于大语言模型的警报分类
From Investigation Failures to Reliable SOC Agents: Understanding and Improving LLM-Based Alert Triage
arXiv 2610.10608 · 2026-10-09相机原生立体VR180数据集
A Camera-Native Stereo VR180 Dataset
arXiv 2610.10607 · 2026-10-09感知肤色的性别化影响：1870-1940年非裔美国兄弟姐妹的证据
The Gendered Impacts of Perceived Skin Tone: Evidence from African American Siblings in 1870-1940
arXiv 2610.10606 · 2026-10-09时变螺旋对称下的动力学约化与流体动力学
Kinetic Reduction and Hydrodynamics under Time-Dependent Helical Symmetry
arXiv 2610.10605 · 2026-10-09超越类型检查：迈向形式化规范生成的整体评估
Beyond Type-checking: Towards Holistic Evaluation of Formal Specification Generation
arXiv 2610.10604 · 2026-10-09$\mathbb{R}^n$中余维二平面的Kakeya问题
The Kakeya problem for codimension-two planes in $\mathbb{R}^n$
arXiv 2610.10603 · 2026-10-09具有非线性空间位阻的核糖体流动模型中的夹带增益：全局刚性与维度诱导增益
Gain of Entrainment in Ribosome Flow Models with Nonlinear Steric Hindrance: Global Rigidity and Dimension-Induced Gai
arXiv 2610.10602 · 2026-10-09教机器狗新技能：结合强化学习与模仿学习及对抗性任务选择实现四足机器人多样技能
Teaching a Robot Dog New Tricks: Diverse Quadruped Skills via Combined Reinforcement and Imitation Learning with Adversarial Task Selection
arXiv 2610.10601 · 2026-10-09VC学习的最优信息复杂度
The optimal information complexity of VC learning
arXiv 2610.10600 · 2026-10-09随机3-SAT可满足性阈值的一个上界：4.268
An upper bound of 4.268 for the random 3-SAT satisfiability threshold
arXiv 2610.10599 · 2026-10-09全双工大规模MIMO ISCC网络的性能分析与感知感知资源分配
Performance Analysis and Sensing-Aware Resource Allocation for Full-Duplex Massive MIMO ISCC Networks
arXiv 2610.10598 · 2026-10-09可验证腐败预算：自适应操纵下的任意时刻有效排行榜声明
Certified Corruption Budgets: Anytime-Valid Leaderboard Claims under Adaptive Rigging
arXiv 2610.10597 · 2026-10-09Dijkstra算法并非贪心算法：一种全局到局部的正确性证明
Dijkstra Is NOT Greedy: A Global-to-Local Proof of Correctness
arXiv 2610.10596 · 2026-10-09Dirichlet L-函数的简单且精确近似
Simple and accurate approximations to Dirichlet $L$-functions
arXiv 2610.10595 · 2026-10-09覆盖度而非难度决定激活探针所需合成数据量
Coverage, Not Difficulty, Sets How Much Synthetic Data an Activation Probe Needs
arXiv 2610.10594 · 2026-10-09通过结构化优化实现更快的矩阵乘法算法
A faster matrix multiplication algorithm through structured optimization
arXiv 2610.10593 · 2026-10-09Wieszcz-XIX：一个包含31亿词的1918年前波兰语语料库及从头训练的时间受限语言模型
Wieszcz-XIX: A 3.1-Billion-Word Corpus of Pre-1918 Polish and Temporally Bounded Language Models Trained From Scratch
arXiv 2610.10592 · 2026-10-09马尔可夫可达拓扑及其概率结构
The Markov Accessibility Topology and its Probabilistic Structure
arXiv 2610.10591 · 2026-10-09工具使用智能体的智能体控制遗忘：实践中的可逆上下文管理
Agent-Controlled Forgetting for Tool-Using Agents: Reversible Context Curation in Practice
arXiv 2610.10590 · 2026-10-09中子星内部的普朗克尺度黑洞残余
Planck-scale black hole remnants inside neutron stars
arXiv 2610.10589 · 2026-10-09当量子场理论介入时：物质、引力与能量之间的关系
When quantum field theory kicks in: the relation between matter, gravity and energy
arXiv 2610.10588 · 2026-10-09《太空符号》第二部分：一个新的模拟首次接触场景，吸引全球SETI科学网络参与，并将深空技术整合到SETI搜索中
A Sign in Space part 2, a new simulated First Contact scenario engaging the global scientific SETI network, and integrating deep space technologies into SETI search
arXiv 2610.10587 · 2026-10-09群的局部表示：完备化与史密斯猜想
Local Presentations of Groups: Completion and Smith's Conjecture
arXiv 2610.10586 · 2026-10-09动力学暗能量与非平坦宇宙学中的部分热化电子伏特量级惰性中微子
Partially Thermalized eV-Scale Sterile Neutrinos in Dynamical Dark Energy and Non-Flat Cosmologies
arXiv 2610.10585 · 2026-10-09Balazard--Saias--Yor准则的有限半径Jensen与相对熵形式
Finite-radius Jensen and relative-entropy formulations of the Balazard--Saias--Yor criterion
arXiv 2610.10584 · 2026-10-09非阿基米德局部域上局部ε因子的朗兰兹第一主引理的一个局部证明
A Local Proof of Langlands's First Main Lemma for Local Epsilon Factors over Nonarchimedean Local Fields
arXiv 2610.10583 · 2026-10-09素数间隙中粗数的重数常数
Multiplicity Constants for Rough Numbers in Prime Gaps
arXiv 2610.10582 · 2026-10-09高斯超几何函数的第一类幂平均界
The first type of power mean bounds for the Gaussian hypergeometric function
arXiv 2610.10581 · 2026-10-09集成大语言模型（LLM）的硬件设计验证综述
A Survey on LLM-Integrated Hardware Design Verification
arXiv 2610.10580 · 2026-10-09CacheFit：末级缓存替换策略的缓存替换成本分析与约束驱动式重新设计
CacheFit: The Rising Cost of Cache Replacement and Constraint-Driven Redesign of Last-Level Cache Replacement Policies
arXiv 2610.10579 · 2026-10-09Rogers-Ramanujan连分数的对数矩与奇数zeta值
Logarithmic moments of the Rogers Ramanujan continued fraction and odd zeta values
arXiv 2610.10578 · 2026-10-09映射类群$\boldsymbol{\rm Mod}(S_g)$在$(2g-1)$维球面上的忠实作用
A Faithful Action of the Mapping Class Group $\mathrm{Mod}(S_g)$ on the $(2g-1)$-Sphere
arXiv 2610.10577 · 2026-10-09不可约尖点eta商的结构
The structure of irreducible cuspidal eta quotients
arXiv 2610.10576 · 2026-10-09$L^2$ 至7维标量曲率有界的Ricci流的曲率界
$L^2$ Curvature Bounds on Ricci flows with bounded scalar curvature up to dimension 7
arXiv 2610.10575 · 2026-10-09一类涉及多元q-米塔格-莱夫勒函数的新型q-分数阶积分算子
On a new class of $q$-fractional integral operators involving multivariate $q$-Mittag-Leffler functions
arXiv 2610.10574 · 2026-10-09何种形式的学生意义建构可预测数学学习？：学生对机会、舒适感和效用的视角
What Forms of Student Sensemaking Predict Mathematics Learning?: Student Perspectives on Opportunity, Comfort, and Utility
arXiv 2610.10573 · 2026-10-09∑_{i=1}^{n}1/gcd(n,i) 的整数值
The Integer Values of $\sum_{i=1}^{n}\frac{1}{\gcd(n,i)}$
arXiv 2610.10572 · 2026-10-09SPERA：具有几何与频率感知隐空间预测的球形先验脑电图基础模型
SPERA: Spherical Prior EEG Foundation Model with Geometry- and Frequency-Aware Latent Prediction
arXiv 2610.10571 · 2026-10-09
