2026-10论文中文摘要
GeoGAT：双向时间采样结合层次图注意力用于全球视频地理定位
GeoGAT: Bidirectional Temporal Sampling Meets Hierarchical Graph Attention for Global Video Geo-localization
arXiv 2609.39128 · 2026-10-01如何减少定位歧义？用于卫星-地面定位的几何-语义约束BEV表示学习
How to Reduce Localization Ambiguity? Geometry-Semantic Constrained BEV Representation Learning for Satellite-Ground Localization
arXiv 2609.39127 · 2026-10-01纠缠的不确定性原理
An uncertainty principle for entanglement
arXiv 2609.39126 · 2026-10-01LBDU-VIO：面向视觉不可靠场景的基于学习偏置动力学与不确定性的视觉惯性里程计
LBDU-VIO: Learned Bias Dynamics and Uncertainty for Visual-Inertial Odometry with Unreliable Vision
arXiv 2609.39125 · 2026-10-01CDMD：用于表格数据的跨数据集混合类型扩散模型
CDMD: A Cross-Dataset Mixed-Type Diffusion Model for Tabular Data
arXiv 2609.39124 · 2026-10-01用于所选GP近最优性证书的先验E值
Prequential E-Values for Selected-GP Near-Optimality Certificates
arXiv 2609.39123 · 2026-10-01Coulomb 分支的对偶函子 I
Duality functors for Coulomb branches I
arXiv 2609.39122 · 2026-10-01更好的教师监督是否足够？多模态在线策略蒸馏中学生侧学习的解锁
Is Better Teacher Supervision Enough? Unlocking Student-side Learning in Multimodal On-Policy Distillation
arXiv 2609.39120 · 2026-10-01诊断推理语言模型的同策略自蒸馏
Diagnosing On-Policy Self-Distillation for Reasoning Language Models
arXiv 2609.39118 · 2026-10-01SteerProbe：学习绕过视觉-语言模型中的安全转向
SteerProbe: Learning to Bypass Safety Steering in Vision-Language Models
arXiv 2609.39117 · 2026-10-01GRC-Pose：用于无先验6D物体姿态跟踪的生成-重建对应
GRC-Pose: Generation-Reconstruction Correspondence for Prior-Free 6D Object Pose Tracking
arXiv 2609.39116 · 2026-10-01超越局部线性：学习型图像编码器的尺度分辨几何
Beyond Local Linearity: Scale-Resolved Geometry of Learned Image Encoders
arXiv 2609.39115 · 2026-10-01费米子和玻色子高斯态的最优测试
Optimal testing of fermionic and bosonic Gaussian states
arXiv 2609.39113 · 2026-10-01CamAgent：用于多物种相机陷阱工作流的LLM智能体框架
CamAgent: An LLM-Agent Framework for Multi-Species Camera-Trap Workflows
arXiv 2609.39112 · 2026-10-01Bongard：训练机器直觉
Bongard: Training Machine Intuition
arXiv 2609.39111 · 2026-10-01施瓦西黑洞波光学成像中两个邻近非相干源的四极响应
Second-order multipole response of two nearby incoherent sources in wave-optical imaging of a Schwarzschild black hole
arXiv 2609.39110 · 2026-10-01T-Router：通过参数高效强化学习学习丘脑路由进行推理
T-Router: Learning Thalamic Routing for Reasoning with Parameter-Efficient Reinforcement Learning
arXiv 2609.39109 · 2026-10-01交变磁性α-MnTe中175-180 cm$^{-1}$拉曼特征的本质
Nature of the 175-180 cm$^{-1}$ Raman Feature in Altermagnetic $α$-MnTe
arXiv 2609.39108 · 2026-10-01MASCRDM：大型语言模型训练过程中合规风险检测与缓解的多智能体系统
MASCRDM: Multi-Agent System for Compliance Risk Detection and Mitigation in Training Process of Large Language Models
arXiv 2609.39107 · 2026-10-01面向不确定基线的保守型多臂赌博机：考虑储备的对比证书
Reserve-Aware Contrast Certificates for Conservative Bandits with Uncertain Baselines
arXiv 2609.39106 · 2026-10-01遮挡感知、准静态、面向稳定性的不平地形轨迹规划
Occlusion-Aware, Quasi-Static, Stability-Oriented Trajectory Planning on Uneven Terrain
arXiv 2609.39105 · 2026-10-01加权右角Artin群的拟射影性
Quasi-Projectivity of Weighted Right-Angled Artin Groups
arXiv 2609.39104 · 2026-10-01关于永恒连通顶点覆盖
On Eternal Connected Vertex Cover
arXiv 2609.39103 · 2026-10-01超越预测：利用回顾性世界建模引导VLM智能体
Beyond Prediction: Steering VLM Agents with Retrospective World Modeling
arXiv 2609.39101 · 2026-10-01西蒙斯天文台：大口径望远镜探测器读出系统的设计与初始性能
The Simons Observatory: Design and Initial Performance of the Detector Readout System for the Large-Aperture Telescope
arXiv 2609.39100 · 2026-10-01泛化信号未必是模型选择信号
A Generalisation Signal Need Not Be a Model-Selection Signal
arXiv 2609.39099 · 2026-10-01DiFF：面向人体运动流的多普勒信息流匹配
DiFF: Doppler-informed Flow Matching for Human Motion Flow
arXiv 2609.39098 · 2026-10-01具有可选偏好的桥梁选址机制设计
Mechanism Design for Bridge Location with Optional Preferences
arXiv 2609.39097 · 2026-10-01基础权处 $C_n$ 型不等参数 Kostka--Foulkes 多项式
Unequal-parameter Kostka--Foulkes polynomials of type $C_n$ at fundamental weights
arXiv 2609.39095 · 2026-10-01三维拓扑结构：大角度扭转石墨中堆叠控制的Umklapp散射
Three-Dimensional Topology from Stacking-Controlled Umklapp Scattering in Large-Angle Twisted Graphite
arXiv 2609.39094 · 2026-10-01通过状态增强学习无限时域平均奖励约束马尔可夫决策过程
Learning Infinite-Horizon Average-Reward CMDPs via State Augmentation
arXiv 2609.39093 · 2026-10-01量子电路编译的常数开销
Quantum circuit compilation with constant overhead
arXiv 2609.39092 · 2026-10-01原子级机器学习生态系统的战略路线图
A strategic roadmap for an atomistic machine-learning ecosystem
arXiv 2609.39090 · 2026-10-01UGOD：用于稀疏视角3D高斯泼溅的不确定性引导的不透明度与丢弃
UGOD: Uncertainty-Guided Opacity and Dropout for Sparse-View 3D Gaussian Splatting
arXiv 2609.39089 · 2026-10-01SCIC：面向说话人自适应表现力TTS的范围与码本感知指令条件化
SCIC: Scope- and Codebook-Aware Instruction Conditioning for Speaker-Adapted Expressive TTS
arXiv 2609.39088 · 2026-10-01洛伦兹破缺暴胀中非正则动力学下的原初张量信号与引力波约束
Primordial Tensor Signatures and Gravitational Wave Constraints in Lorentz Violating Inflation with Non-Canonical Kinetics
arXiv 2609.39087 · 2026-10-01真实仓库中可信的运行时错误修复：基准与护栏
Trustworthy Runtime Error Healing in Real-World Repositories: A Benchmark and Guardrail
arXiv 2609.39086 · 2026-10-01Kerr-AdS$_5$黑洞的复相结构：临界鞍点与Fisher零点
Complex Phase Structure of Kerr-AdS$_5$ Black Holes: Critical Saddles and Fisher Zeros
arXiv 2609.39085 · 2026-10-01MRI超分辨率与RCDM/WaveMix及任务感知分割
MRI Super-Resolution with RCDM/WaveMix and Task-Aware Segmentation
arXiv 2609.39083 · 2026-10-01编码理论的编码智能体
Coding Agents for Coding Theory
arXiv 2609.39081 · 2026-10-01关联特征条件化的集合-时序Transformer用于跨会话皮层内运动解码
Association profile conditioning in a set-temporal transformer for cross-session intracortical motor decoding
arXiv 2609.39080 · 2026-10-01对称品种环路中轨道闭包的奇点 II：扭曲情形
Singularities of orbit closures in loop spaces of symmetric varieties II: twisted cases
arXiv 2609.39079 · 2026-10-01参数对称性决定过参数化非线性网络中的表征几何
Parameter symmetries determine representational geometry in overparameterized nonlinear networks
arXiv 2609.39078 · 2026-10-01信息甄别的容量成本
The Capacity Cost of Informational Screening
arXiv 2609.39077 · 2026-10-01RAGScope：一种泄漏受控、成本感知的证据门控协议，用于RAG幻觉分诊
RAGScope: A Leakage-Controlled, Cost-Aware Evidence-Gating Protocol for RAG Hallucination Triage
arXiv 2609.39075 · 2026-10-01HO-FL：面向异构边缘设备的混合阶联邦学习
HO-FL: Hybrid-Order Federated Learning for Heterogeneous Edge Devices
arXiv 2609.39074 · 2026-10-01操作条件下强度调制光致发光的载流子动力学：从钙钛矿薄膜到太阳能电池
Operando Charge Carrier Dynamics by Intensity-Modulated Photoluminescence: From Perovskite Thin Films to Solar Cells
arXiv 2609.39073 · 2026-10-01超越文本：基于LLM的多模态对话中的维度情感评估
Beyond Text: LLM-Based Dimensional Emotion Evaluation in Multimodal Dialogue
arXiv 2609.39072 · 2026-10-01LexReward：面向法律语言模型的分类学驱动奖励框架
LexReward: A Taxonomy-Driven Reward Framework for Legal Language Models
arXiv 2609.39071 · 2026-10-01双刚性TANK模型中的继承工资离散与最优自由裁量
Inherited Wage Dispersion and Optimal Discretion in a Dual-Rigidity TANK Model
arXiv 2609.39070 · 2026-10-01
