2026-08论文中文摘要
DynaKRAG：多跳检索增强生成中可学习证据控制的统一框架
DynaKRAG: A Unified Framework for Learnable Evidence Control in Multi-Hop Retrieval-Augmented Generation
arXiv 2607.06507 · 2026-08-25李代数子代数的收敛实现
Convergent realizations of Lie subalgebras
arXiv 2607.06490 · 2026-08-25FSDC-DETR：面向小目标检测的频域-空域协同DETR
FSDC-DETR: A Frequency-Spatial Domain Collaborative DETR for Small Object Detection
arXiv 2607.05176 · 2026-08-25关于联合传递的可交换极小同胚
On jointly transitive commuting minimal homeomorphisms
arXiv 2607.05161 · 2026-08-25铁磁平面腔中的有限频率磁性公共浴场
Finite-frequency magnetic common baths in ferromagnetic planar cavities
arXiv 2607.05066 · 2026-08-25Athena-WBC：用于长尾人形机器人全身控制的能力对齐策略专家
Athena-WBC: Capability-Aligned Policy Experts for Long-Tail Humanoid Whole-Body Control
arXiv 2607.04837 · 2026-08-25σ有效性的分类
Classification of $σ$-validity in iterated announcements
arXiv 2607.04685 · 2026-08-25二维α-石墨二炔电子结构与光学性质的第一性原理研究
Structural, electronic, and optical properties of 2D $α$-graphdiyne from first-principles
arXiv 2607.03841 · 2026-08-25通过同构-异构分割对合成图像进行生成后筛选
Post-Generation Curation of Synthetic Images via Homogeneous-Heterogeneous Splitting
arXiv 2607.02637 · 2026-08-25H-OPD：置信感知异构多教师多模态在线策略蒸馏
H-OPD: Confidence Aware Heterogeneous Multi-Teacher Multimodal On-policy Distillation
arXiv 2607.02592 · 2026-08-25GeoMix: 通过全局上下文和多检测器训练实现无描述符视觉定位
GeoMix: Descriptor-Free Visual Localization via Global Context and Multi-Detector Training
arXiv 2607.02486 · 2026-08-25相对曲线上齐性空间的梳子光滑化与局部平凡性
Comb smoothing and local triviality of homogeneous spaces over a relative curve
arXiv 2607.02014 · 2026-08-25PFLOTRAN中顺序热-孔隙弹性公式的验证
Verification of a sequential thermo-poroelasticity formulation in PFLOTRAN
arXiv 2607.01156 · 2026-08-25准德西特背景的隐藏量子信息对称性
Hidden quantum-informatic symmetries of quasi-de Sitter backgrounds
arXiv 2607.00636 · 2026-08-25群等变庞加莱卷积网络
Group-Equivariant Poincaré Convolutional Networks
arXiv 2607.00556 · 2026-08-25基于相位分解强化学习的分布式多机器人月球货物运输
Distributed Multi Robot Lunar Cargo Transportation via Phase Decomposed Reinforcement Learning
arXiv 2607.00160 · 2026-08-25用于探测真空非线性的经典与非经典光子态
Classical versus non-classical photon states for detecting vacuum non-linearity
arXiv 2608.21361 · 2026-08-24OmniAssistBench：面向全模态大语言模型（Omni-LLMs）的助手式交互基准
OmniAssistBench: Assistant-style Interaction Benchmark for Omni-LLMs
arXiv 2608.21360 · 2026-08-24牛顿法的原始加速
Primal Acceleration of Newton's Method
arXiv 2608.21359 · 2026-08-24利用太空机器人开展地外采矿：探测、采样与提取
Mining beyond Earth with Space Robots: Exploration, Sampling, and Extraction
arXiv 2608.21358 · 2026-08-24ViTacPhys：基于人类视觉-触觉演示的感知物体物理属性的抓取方法
ViTacPhys: Physical Property-Aware Grasping from Human Visual-Tactile Demonstrations
arXiv 2608.21355 · 2026-08-24高能中微子对受记忆负荷的巴丁（Bardeen）和克尔（Kerr）原初黑洞的约束
High-Energy Neutrino Constraints on memory burdened Bardeen and Kerr Primordial Black Holes
arXiv 2608.21353 · 2026-08-24纤维收缩扩张的上同调约化：从子上同调到热力学形式主义
Cohomological Reduction for Fiber-Contracting Extensions:From Subcohomology to Thermodynamic Formalism
arXiv 2608.21352 · 2026-08-24标准排列中的模式规避
Pattern avoidance in canon permutations
arXiv 2608.21351 · 2026-08-24原子干涉测量中的魔术速度选择
Magic Velocity Selection in Atom Interferometry
arXiv 2608.21350 · 2026-08-24PerturbRx：用于患者药物反应预测的治疗条件潜在迁移学习
PerturbRx: Learning Treatment-Conditioned Latent Transitions for Patient Drug Response Prediction
arXiv 2608.21349 · 2026-08-24序列预测的真实校准度量
Truthful Calibration Measures for Sequential Prediction
arXiv 2608.21348 · 2026-08-24超图完美匹配的精确最小$d$度阈值
Exact Minimum $d$-Degree Thresholds for Hypergraph Perfect Matchings
arXiv 2608.21347 · 2026-08-24素幂模下的Kloosterman和的乘积和
Sums of products of Kloosterman sums to prime power moduli
arXiv 2608.21346 · 2026-08-24自精化流水线中的非对称容量分配
Asymmetric Capacity Allocation in Self-Refinement Pipelines
arXiv 2608.21345 · 2026-08-24RHIC和LHC中sQGP内的底夸克偶素输运
Bottomonium transport in the sQGP at RHIC and the LHC
arXiv 2608.21344 · 2026-08-24TurboBias 2.0：面向生产高效型ASR系统的流式上下文偏置技术
TurboBias 2.0: Streaming Context-Biasing for Production-Efficient ASR Systems
arXiv 2608.21343 · 2026-08-24虚拟蒸馏成功概率的斯莱皮恩界
Slepian Bounds on the Success Probability of Virtual Distillation
arXiv 2608.21342 · 2026-08-24自然语言工作流尚未成为软件：面向可靠智能体执行的工件驱动编译
Natural-Language Workflows Are Not Software Yet: Artifact-Driven Compilation for Reliable Agent Execution
arXiv 2608.21341 · 2026-08-24当s接近0时，三维欧氏空间中的稳定s-极小锥是平坦的
Stable $s$-minimal cones in $\mathbb{R}^3$ are flat for $s$ close to zero
arXiv 2608.21340 · 2026-08-24有限域上斜常循环码的首个紧分类
The first tight classification of skew-constacyclic codes over finite fields
arXiv 2608.21339 · 2026-08-24面向高多重度LHC过程的高效事例生成：基于归一化流的端到端GPU工作流
Efficient Event Generation for High-Multiplicity LHC Processes: An End-to-End GPU Workflow with Normalizing Flows
arXiv 2608.21338 · 2026-08-24马尔可夫链蒙特卡洛中的麦克斯韦妖：冷却信息流与熵平衡
Maxwell's Demon in Markov Chain Monte Carlo: Cooling Information Flow and Entropy Balance
arXiv 2608.21337 · 2026-08-24非相互作用量子链子系统动力学的耗散框架
Dissipative framework for subsystem dynamics of noninteracting quantum chains
arXiv 2608.21336 · 2026-08-24巴拿赫空间上有界算子构成的巴拿赫代数的极大右理想
Maximal right ideals of the Banach algebra of bounded operators on a Banach space
arXiv 2608.21335 · 2026-08-24用于高性能隧穿场效应晶体管的拓扑半导体Mo₂SeTe₃中的狄拉克表面态与非局域量子隧穿
Dirac Surface States and Nonlocal Quantum Tunneling in Topological Semiconductor Mo$_2$SeTe$_3$ for High-Performance Tunnel FETs
arXiv 2608.21333 · 2026-08-24解剖学先验引导神经网络：在损失函数与架构中编码解剖学先验，并结合导丝诱导主髂动脉变形的SE(3)公式
Anatomy-Informed Neural Networks: Encoding Anatomic Priors in Loss and Architecture, with an SE(3) Formulation of Guidewire-Induced Aortoiliac Deformation
arXiv 2608.21332 · 2026-08-24具有准周期非线性的三个核心耦合波导中局域解的动力学
Dynamics of localized solutions in three core coupled waveguides with quasi-periodic nonlinearity
arXiv 2608.21331 · 2026-08-24NeSAM：用于越野移动的带土壤自适应的神经符号运动动力学模型
NeSAM: Neuro-Symbolic Kinodynamics with Soil Adaptation for Off-Road Mobility
arXiv 2608.21330 · 2026-08-24通过一阶相变的量子退火：场论方法
Quantum annealing through a first-order phase transition: field theory approach
arXiv 2608.21329 · 2026-08-24通过Stratonovich-Weyl对应从量子场论得到经典动力学
Classical dynamics from QFT via Stratonovich-Weyl correspondence
arXiv 2608.21328 · 2026-08-24隐形智能体，不知情患者：撒哈拉以南非洲自主AI诊断智能体的负责任部署
Invisible Agents, Uninformed Patients: Towards Responsible Deployment Of Autonomous AI Diagnostic Agents In Sub-Saharan Africa
arXiv 2608.21326 · 2026-08-24逐步推进：测量与引导大语言模型开展心理治疗的方式
Move by Move: Measuring and Steering How LLMs Conduct Psychotherapy
arXiv 2608.21325 · 2026-08-24基于时间感知Transformer的AECOPD预测模型
Time-Aware Tranformer-Based Prediction Model for AECOPD
arXiv 2608.21324 · 2026-08-24来自小尺度等曲率的星系中心多样性
Diversity of Galaxy Centers from Small-Scale Isocurvature
arXiv 2608.21323 · 2026-08-24
