2026-09论文中文摘要
SceneBench：面向3D场景视觉-语言理解的分层基准
SceneBench: A Hierarchical Benchmark for Vision-Language Understanding of 3D Scenes
arXiv 2609.16233 · 2026-09-16面向治理感知的自主GIS：LLM赋能的GeoAI中伦理与隐私风险综述
Toward Governance-Aware Autonomous GIS: A Narrative Review of Ethical and Privacy Risks in LLM-Enabled GeoAI
arXiv 2609.16232 · 2026-09-16RuleAutoPilot：从网络流量综合生成可部署的 Suricata 规则
RuleAutoPilot: Synthesizing Deployable Suricata Rules from Network Traffic
arXiv 2609.16231 · 2026-09-16有限度量空间上的 Lipschitz-free 空间中 $\ell_\infty^n$ 的等距拷贝
Isometric copies of $\ell_\infty^n$ in Lipschitz-free spaces over finite metric spaces
arXiv 2609.16230 · 2026-09-16基于稀疏自编码器的测试时机器遗忘
Test-Time Unlearning via Sparse Autoencoder
arXiv 2609.16229 · 2026-09-16教吸血鬼新把戏：神经子句选择的实验研究
Teaching Vampire New Tricks: An Experimental Study of Neural Clause Selection
arXiv 2609.16228 · 2026-09-16互素作用与不动点子群上非零的特征标
Coprime Actions and Characters Non-vanishing on the Fixed-point Subgroup
arXiv 2609.16227 · 2026-09-16信息性抽样下基于样本级权重模型的相干且稳健贝叶斯推断
Coherent and robust Bayesian inference under informative sampling via sample-level weight models
arXiv 2609.16225 · 2026-09-16太阳系能教会我们关于岩石系外行星的什么
What the Solar System Can Teach Us About Rocky Exoplanets
arXiv 2609.16224 · 2026-09-16空间图的运动规划与拓扑复杂度
Motion planning and topological complexity for diagrams of spaces
arXiv 2609.16223 · 2026-09-16我如何学会不再担忧并爱上StopGrads：平稳性、收敛性及流映射学习案例研究
How I learned to stop worrying and love StopGrads: Stationarity, Convergence, and a case study on Flow Map Learning
arXiv 2609.16222 · 2026-09-16向量场、初始支架与数据库约简
Vector fields, initial scaffolds and database reduction
arXiv 2609.16221 · 2026-09-16无穷远处的压力与大偏差
Pressure at infinity and Large Deviations
arXiv 2609.16220 · 2026-09-16QCD企鹅算符对粲夸克CP破坏的标准模型上界的增强
Enhanced Standard Model bound on charm CP violation from QCD penguins
arXiv 2609.16219 · 2026-09-16如何缩小测试数据库族？带空值和比较的查询包含问题
How Can We Shrink the Family of Test Databases? Query Containment with Nulls and Comparisons
arXiv 2609.16218 · 2026-09-16一种神经-星形胶质细胞架构实现用于证据累积的混合自动机
A neural-astrocyte architecture implements a hybrid automaton for evidence accumulation
arXiv 2609.16217 · 2026-09-16KV 缓存应存放于何处？面向长生命周期会话的 GPU、CPU 与 SSD 放置策略
Where Should the KV Cache Live? Placement Policies Across GPU, CPU, and SSD for Long-Lived Sessions
arXiv 2609.16215 · 2026-09-16人工智能与生物安全：能力、威胁路径与纵深防御治理
Artificial intelligence and biosecurity: capabilities, threat pathways, and defense-in-depth governance
arXiv 2609.16213 · 2026-09-16低体积双曲3-轨形的基础流形
The underlying manifold of a low-volume hyperbolic 3-orbifold
arXiv 2609.16212 · 2026-09-16基因组基础模型同态推理的可行性
Feasibility of Homomorphic Inference for a Genomic Foundation Model
arXiv 2609.16211 · 2026-09-16机器学习群体合成校准并合率在IllustrisTNG300中的引力波偏置
Gravitational Wave Bias in IllustrisTNG300 from Machine-Learned Population-Synthesis Calibrated Merger Rates
arXiv 2609.16209 · 2026-09-16用于空间转录组学的双曲对比学习与蕴含关系
Hyperbolic Contrastive Learning with Entailment for Spatial Transcriptomics
arXiv 2609.16207 · 2026-09-16校准，然后路由：面向分离式LLM服务的学习型请求路由的实测研究
Calibrate, Then Route: A Measured Study of Learned Request Routing for Disaggregated LLM Serving
arXiv 2609.16206 · 2026-09-16碰撞诱导磁重联形成纤维结构——运动学特征
Filament Formation via Collision-induced Magnetic Reconnection -- Kinematic Features
arXiv 2609.16205 · 2026-09-16诱饵方向优化：针对LLM消融的后置防御
Decoy Direction Optimization: A Post-Hoc Defense Against LLM Abliteration
arXiv 2609.16204 · 2026-09-16官方贫困线之外的全球贫困：物质匮乏的有界估计
Global Poverty Beyond the Official Line: A bounded estimate of material insufficiency
arXiv 2609.16203 · 2026-09-16非交换 $L_p$-空间中的锐切向不等式
Sharp tangent inequalities in noncommutative $L_p$-spaces
arXiv 2609.16202 · 2026-09-16拯救开普勒的假阳性：三合星系统KOI-1623中一颗可能的宜居带系外行星
Rescuing Kepler's False Positives: A Possible Habitable Zone Exoplanet in the Triple Star System KOI-1623
arXiv 2609.16201 · 2026-09-16一个用于深度学习活动火灾分割的哨兵二号基准数据集：覆盖加利福尼亚州25起野火
A Sentinel-2 benchmark dataset for deep-learning active-fire segmentation across 25 California wildfires
arXiv 2609.16199 · 2026-09-16基于分层最短路径的量子网络效用路径选择与配置
Utility-Based Path Selection and Configuration in Quantum Networks via Layered Shortest Paths
arXiv 2609.16198 · 2026-09-16通过核束穿过非极化靶时的自旋过滤获得自旋 $S\ge1$ 的张量极化核（自旋二色性效应）
Obtaining tensor-polarized nuclei with spin $S\ge1$ via spin filtering upon nuclear beam transmission through an unpolarized target (spin dichroism effect)
arXiv 2609.16197 · 2026-09-16局部上同调的滤过与单射性定理
Filtrations on Local Cohomology and Injectivity Theorems
arXiv 2609.16196 · 2026-09-16多胞体的组合切片问题：如何（不）从切片重构多胞体
Combinatorial slicing problems of polytopes: How (not) to reconstruct a polytope from its slices
arXiv 2609.16195 · 2026-09-16透过最后散射面之纱：利用CMB谱畸变观测宇宙
Peering Beyond the Veil of Last Scattering: A View of the Universe with CMB Spectral Distortions
arXiv 2609.16194 · 2026-09-16基于置换的大型语言模型隐恶意软件：威胁与对策
Permutation-Based Stegomalware in Large Language Models: Threats and Countermeasures
arXiv 2609.16193 · 2026-09-16AI驱动的反馈系统、数字劳动与静默离职：变革非洲职场
AI-Driven Feedback Systems, Digital Labour, and Silent Quitting: Transforming African Workplaces
arXiv 2609.16192 · 2026-09-16当AI说“我无法回答”：理解用户对AI拒绝回应的反应
When AI Says "I Am Unable to Answer": Understanding User Responses to AI Refusals
arXiv 2609.16191 · 2026-09-16JWST对星暴的观测：NGC 253中心星暴中的一个年轻气泡
JWST Observations of Starbursts: A Young Bubble in NGC 253's Central Starburst
arXiv 2609.16190 · 2026-09-16观点：人工智能尚未准备好应对战略冲突
Position: AI Is Not Ready for Strategic Conflicts
arXiv 2609.16189 · 2026-09-16两类扭曲里德-所罗门码的变体
Two variants of Twisted Reed-Solomon Codes
arXiv 2609.16188 · 2026-09-16淬火锻造：EAGLE 模拟中恒星形成星系与静止星系间的形态转变
Forged in Quenching: Morphological Transformation across Star-forming and Quiescent Galaxies in EAGLE
arXiv 2609.16187 · 2026-09-16占用网络引导的自主机器人肾部分切除术
Occupancy Network-Guided Autonomous Robotic Partial Nephrectomy
arXiv 2609.16186 · 2026-09-16面向连续谐波负荷建模的分布式边缘到云架构
Distributed Edge-to-Cloud Architecture for Continual Harmonic Load Modeling
arXiv 2609.16185 · 2026-09-16Powers-Størmer-Ogata 不等式的一个推广
A generalization of the Powers-Størmer-Ogata inequality
arXiv 2609.16184 · 2026-09-16固定状态递归中联想记忆的解剖：匹配状态分解、干扰墙与打破它的课程
Anatomy of Associative Recall in Fixed-State Recurrences: A Matched-State Decomposition, an Interference Wall, and a Curriculum That Breaks It
arXiv 2609.16183 · 2026-09-16线性规划的GPU加速预求解
GPU-Accelerated Presolving for Linear Programming
arXiv 2609.16182 · 2026-09-16Mean Lindelöf 假设与 Bianchi 流形上的素数测地线定理
Mean Lindelöf Hypothesis and Prime Geodesic Theorem on Bianchi Manifolds
arXiv 2609.16181 · 2026-09-16超光速物体的延迟光渲染
Delayed-Light Rendering for Superluminal Objects
arXiv 2609.16180 · 2026-09-16Z-Loss在稠密输出头和稀疏路由器中的反向几何
Z-Loss Backward Geometry in Dense Output Heads and Sparse Routers
arXiv 2609.16179 · 2026-09-16紧致连通李群的Mathieu性质
The Mathieu Property for Compact Connected Lie Groups
arXiv 2609.16178 · 2026-09-16
