2026-09论文中文摘要
图匹配松弛与摊销用于监督图预测
Graph Matching Relaxations and Amortization for Supervised Graph Prediction
arXiv 2609.15437 · 2026-09-28解释说话人嵌入的层次组织
Interpreting hierarchical organisation of speaker embeddings
arXiv 2609.15203 · 2026-09-28T-LoopFormer：用于潜在推理的动态路由令牌级弹性深度循环Transformer
T-LoopFormer: Token-Level Elastic-Depth Looped Transformers for Latent Reasoning with Dynamic Routing
arXiv 2609.15160 · 2026-09-28平面面片和高雅可比双细分六面体模板
Planar-faced and high-Jacobian two-refinement hexahedral templates
arXiv 2609.14729 · 2026-09-28GRAINE 2023乳胶望远镜在亚GeV波段搜寻银河系中心过量辐射的初步结果
First results on the search for the Galactic Center Excess in the sub-GeV band with the emulsion telescope in GRAINE 2023
arXiv 2609.14551 · 2026-09-28对称算子空间的UMD性质
The UMD property of symmetric operator spaces
arXiv 2609.14282 · 2026-09-28透明身份验证方法：基于MPC与高效凭证状态处理
Transparent Identity Verification Approach Using MPC and Efficient Credential Status Handling
arXiv 2609.14195 · 2026-09-28局域性的代价：Schwinger-Keldysh有效场论中的Maxwell-Cattaneo电荷输运
The price of locality: Maxwell-Cattaneo charge transport in Schwinger-Keldysh effective field theory
arXiv 2609.13378 · 2026-09-28圆上战略设施选址的随机化 $\frac32$ 近似算法
A Randomized $\frac32$-Approximation for Strategic Facility Location on a Circle
arXiv 2609.12792 · 2026-09-28多核子转移反应中自旋分布的微观描述
Microscopic description of spin distributions in multinucleon tranfer reactions
arXiv 2609.12739 · 2026-09-28四元数二次相位Dunkl变换：其性质与不确定性原理
Quaternion Quadratic Phase Dunkl Transform: Its Properties and Uncertainty Principle
arXiv 2609.12516 · 2026-09-28保持正性：Schwinger-Keldysh有效场论中非高斯噪声的界
Staying positive: bounds for non-Gaussian noise in Schwinger-Keldysh effective field theory
arXiv 2609.12050 · 2026-09-28有限非阿贝尔对称性的广义离散规范化的不可逆选择规则
Non-invertible Selection Rules from Generalized Discrete Gauging of Finite Non-Abelian Symmetries
arXiv 2609.11895 · 2026-09-28调和$(n-1,1)$-型的一个消失结果
A vanishing result for harmonic $(n-1,1)$-forms
arXiv 2609.11665 · 2026-09-28探索说话人识别中的二阶模式识别
Exploring Second-Order Pattern Recognition in Speaker Recognition
arXiv 2609.11182 · 2026-09-28EGGROLL，展开：理解并改进大规模低秩进化策略
EGGROLL, Unrolled: Understanding and Improving Low-Rank Evolution Strategies at Scale
arXiv 2609.10980 · 2026-09-28Cu嵌入SiO$_x$忆阻器件中开关多样性物理起源的研究
On the physical origins of switching diversity in Cu-embedded SiO$_x$ memristive devices
arXiv 2609.10799 · 2026-09-28Rockafellar约束规格下极大单调算子的非极大和
Nonmaximal sums of maximally monotone operators under Rockafellar's constraint qualification
arXiv 2609.10487 · 2026-09-28激子网络计算表达性中的相干优势
Coherent advantage in the computational expressivity of excitonic networks
arXiv 2609.10448 · 2026-09-28Mockenhaupt的三项Hardy-Littlewood优函数猜想
Mockenhaupt's Three-Term Hardy-Littlewood Majorant Conjecture
arXiv 2609.09740 · 2026-09-28特征叠加中线性可达性的高概率保证
High-probability guarantees for linear accessibility in feature superposition
arXiv 2609.09556 · 2026-09-28存在不完美参考标准时的半参数接收者操作特征分析：基于Box-Cox密度比模型
Semiparametric Receiver Operating Characteristic Analysis in the Presence of an Imperfect Reference Standard via a Box-Cox Density Ratio Model
arXiv 2609.09401 · 2026-09-28光谱框架检验小红点中的准恒星假说 I：通过其超爱丁顿光度比衡量LRD——没有超大质量黑洞的迹象
A Spectral Framework for Testing the Quasi-Star Hypothesis in Little Red Dots I: Weighing LRDs by Their Super-Eddington Luminosity Ratios---No Signs of Overmassive Black Holes
arXiv 2609.09265 · 2026-09-28超越智能体工具链：多智能体系统的跨基底权限
From Evidence to Effect: Authority Semantics and Runtime Infrastructure for Stateful Agents
arXiv 2609.08472 · 2026-09-28近极端高曲率黑洞熵的对数修正
Logarithmic correction to the entropy of near-extremal higher-curvature black holes
arXiv 2609.07801 · 2026-09-28扫描缰绳：AI 编码智能体配置中供应链缺陷的实证研究
Scanning the Harness: Configuration Exposures in AI Coding-Agent Supply Chains
arXiv 2609.07360 · 2026-09-28电子束驱动的集体自混合激子极化激元
Electron beam driven collective self hybridized exciton polaritons
arXiv 2609.07164 · 2026-09-28干预干扰反映模型的默认倾向，而非行为方向
Steering Interference Reflects the Model's Defaults, Not the Behavior Directions
arXiv 2609.06951 · 2026-09-28拒绝的几何学：为何事后安全训练脆弱而预训练期间的安全持久
The Geometry of Refusal: Why Post-Hoc Safety Is Fragile and Pretraining-Time Safety Persists
arXiv 2609.06934 · 2026-09-28超临界区域中分数阶 $p$-Laplacian 的梯度估计
Gradient estimates for the fractional $p$-Laplacian in the superquadratic regime
arXiv 2609.06924 · 2026-09-28Thinking with Cameras: 通过动态视角控制进行主动视觉推理的监控视频理解
Thinking with Cameras: Active Visual Reasoning via Dynamic Viewpoint Control for Surveillance Video Understanding
arXiv 2609.06475 · 2026-09-28PLSR：通过局部化潜在体素扩散实现3D物体的渐进式与局部化超分辨率
PLSR: Progressive and Localized Super-Resolution of 3D Objects via Localized Latent Voxel Diffusion
arXiv 2609.06436 · 2026-09-28切触范畴中立方形式的Cartan演算
Cartan calculus of cubical forms in tangent categories
arXiv 2609.05963 · 2026-09-28SimpleMemVLA：一种简单但有效的视觉-语言-动作模型原生视频记忆
SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models
arXiv 2609.05533 · 2026-09-28早期与晚期：基于DESI DR2的暗能量对比研究
Early against Late: A contrast on dark energy in the light of DESI DR2
arXiv 2609.05410 · 2026-09-28面向主动体积量子架构的二维费米-哈伯德基态能量估计算法的编译
Compiling the 2D Fermi-Hubbard ground-state energy estimation algorithm for active volume quantum architectures
arXiv 2609.05316 · 2026-09-28边缘分布相当不错
The marginal is pretty good
arXiv 2609.05225 · 2026-09-28重新审视加瑟尔猜想
Gatheral's Conjecture Revisited
arXiv 2609.05047 · 2026-09-28WEECFP-SuRGE：带亚结构旋转图距离编码的宽嵌入扩展连接指纹
WEECFP-SuRGE: A Position-Aware Substructure Encoding Method for Molecular Property Prediction
arXiv 2609.04672 · 2026-09-28Motion-Omni：面向口语对话的端到端联合语音与全身动作生成模型
Motion-Omni: End-to-End Joint Speech and Full-Body Motion for Spoken Dialogue
arXiv 2609.04250 · 2026-09-28互易性可使机械网络的学习能力减半
Reciprocity can halve what a mechanical network can learn
arXiv 2609.04169 · 2026-09-28树数据库
Tree Databases
arXiv 2609.03538 · 2026-09-28来自人马座A*和LUX-ZEPLIN核反冲事件的亚TeV单中微子暗物质
Sub-TeV Singlino Dark Matter in light from Sagittarius A$^\ast$ and LUX-ZEPLIN Nuclear-Recoil Event
arXiv 2609.02994 · 2026-09-28BuildOcc：面向建筑能源研究的大语言模型居住者智能体平台
BuildOcc: A Large Language Model Occupant Agent Platform for Building Energy Research
arXiv 2609.02729 · 2026-09-28带装饰旗的量子配置空间的代数研究
Algebraic study of quantum configuration spaces of decorated flags
arXiv 2609.02509 · 2026-09-28从特权目标外观学习跟踪
Learning to Track from Privileged Target Appearances
arXiv 2609.02471 · 2026-09-28机器学习筛选与第一性原理晶格动力学从数据驱动候选搜索中确定NaCa(BH$_4$)$_3$的最稳定结构
Machine-learning screening and first-principles lattice dynamics resolve the most stable structure of NaCa(BH$_4$)$_3$ from a data-driven candidate search
arXiv 2609.02001 · 2026-09-28边染色随机k均匀超图中的彩虹Berge哈密顿性
Rainbow Berge Hamiltonicity in edge-colored random $k$-uniform hypergraphs
arXiv 2609.01989 · 2026-09-28基于系统理论视角的Volterra神经网络学习
Learning with Volterra Neural Networks: A System Theoretic Perspective
arXiv 2609.01928 · 2026-09-28一种量化位点再访问的贪心近邻方法：两种同域分布渡鸦物种的比较
A greedy nearest-neighbor approach to quantify site revisitation: comparing two sympatric raven species
arXiv 2609.01858 · 2026-09-28
