2026-08论文中文摘要
分层自改进：一种面向任务特定可进化智能体控制框架的方法
Hierarchical Self-Improvement: A Framework for Task-Specific Evolvable Agent Harnesses
arXiv 2608.08466 · 2026-08-11无补零空间与非闭值域的对偶性及其对Mazur型算子正则化的影响
Duality of uncomplemented null-space and non-closed range and its impact on the regularization of Mazur-type operators
arXiv 2608.08465 · 2026-08-11强场时空特征对真空正负电子对产生的影响
Effect of Strong Field Space-time Features on Vacuum Pair
arXiv 2608.08464 · 2026-08-11Halpern迭代实现单调变分不等式的$\tilde{\mathcal{O}}(ε^{-1/p})$ $p$阶预言复杂度
Halpern Iteration Achieves $\tilde{\mathcal{O}}(ε^{-1/p})$ $p$th-Order Oracle Complexity for Monotone Variational Inequalities
arXiv 2608.08463 · 2026-08-11ARMOR：通过节点压缩缓解前端瓶颈以加速RTL仿真
ARMOR: Accelerating RTL Simulation by Mitigating the Front-End Bottleneck Using Node Compression
arXiv 2608.08462 · 2026-08-11InstructionCrafter：生成一致且高保真的视觉指令
InstructionCrafter: Generating Consistent and High-Fidelity Visual Instructions
arXiv 2608.08460 · 2026-08-11超越表格：用于关系忠实的文档到数据库构建的Doc2DB-Bench
Beyond Tables: Doc2DB-Bench for Relationally Faithful Document-to-Database Construction
arXiv 2608.08459 · 2026-08-11通过反转热传递函数模拟光学微腔
Simulating optical microcavities by inverting the thermal transfer function
arXiv 2608.08458 · 2026-08-11Lyuksyutov 区域中 Landau-de Gennes 极小元的光滑收敛性
Smooth Convergence of Landau-de Gennes Minimizers in the Lyuksyutov Regime
arXiv 2608.08456 · 2026-08-11广义Freud权、离散Painlevé I hierarchy与Hankel行列式的完全渐近展开
Generalized Freud weight, discrete Painlevé I hierarchy and full asymptotics of Hankel determinants
arXiv 2608.08455 · 2026-08-11是什么阻碍了智能体技能的可复用性？来自13.8万个SKILL.md文件的证据
What Keeps Agent Skills from Being Reusable? Evidence from 138K SKILL.md Files
arXiv 2608.08453 · 2026-08-11揭穿骗局：通过有序推理链正则化检测不断变化的有害聊天对话
Calling the Bluff: Detecting Ever-Shifting Harmful Chat Dialogue via Ordered Reasoning Chain Regularization
arXiv 2608.08451 · 2026-08-11de Rham模空间的算子分层中的切线不连续性
Tangent discontinuity in the oper stratification of de Rham moduli spaces
arXiv 2608.08450 · 2026-08-11超声速与超阿尔文湍流中的慢宇宙射线扩散
Slow Cosmic-Ray Diffusion in Supersonic and Super-Alfvénic Turbulence
arXiv 2608.08449 · 2026-08-11部分电离等离子体中由等离子体团介导的二维磁重联
Plasmoid-Mediated 2D Magnetic Reconnection in Partially Ionized Plasmas
arXiv 2608.08448 · 2026-08-11被遗忘的历史还是时间的考验？从信息检索视角回顾与展望检索增强生成（RAG）
Forgotten History or Test-of-Time? Retrospect and Prospect on RAG from an IR Perspective
arXiv 2608.08445 · 2026-08-11TFDMA相干无源光网络（PON）的载波泄漏抑制与功率差均衡
Carrier Leakage Suppression and Power Difference Equalization for TFDMA Coherent PON
arXiv 2608.08444 · 2026-08-11射影簇的基本形式与无穷小对称性
Fundamental forms and infinitesimal symmetries of projective varieties
arXiv 2608.08442 · 2026-08-11关于一般skein代数表示的Santharoubane猜想的一个反证
A Disproof of Santharoubane's Conjecture on Presentations of Generic Skein Algebras
arXiv 2608.08441 · 2026-08-11MGMCL：结合神经常微分方程的多粒度流形对比学习用于跨主体脑电情感识别
MGMCL: Multi-Granularity Manifold Contrastive Learning With Neural ODEs for Cross-Subject EEG Emotion Recognition
arXiv 2608.08440 · 2026-08-11FSTC-Encoder：用于通用射频感知的特征-空间-时间关联学习
FSTC-Encoder: Feature--Spatial--Temporal Correlation Learning for Generalizable RF Sensing
arXiv 2608.08439 · 2026-08-11树的恰当$\boldsymbol{\{a,b\}}$边赋权
Proper $\{a,b\}$-edge-weightings of trees
arXiv 2608.08438 · 2026-08-11人工智能与研究团队
AI and the Research Team
arXiv 2608.08437 · 2026-08-11FreCast：用于临近降水预报的保相振幅残差扩散雷达回波强度优化方法
FreCast: Refining Radar Echo Intensity via Phase-Preserving Amplitude Residual Diffusion for Precipitation Nowcasting
arXiv 2608.08436 · 2026-08-11深度学习动力学系统发现中线性多步法的核定位与全轨迹泛化
Kernel Localization and Whole-Trajectory Generalization for Linear Multistep Methods in Deep Learning-Based Discovery of Dynamical Systems
arXiv 2608.08435 · 2026-08-11输入问题：量子机器学习的一个永久瓶颈
The Input Problem: A Permanent Bottleneck for Quantum Machine Learning
arXiv 2608.08433 · 2026-08-11BAMU：面向冻结预训练神经语音编解码器的比特流感知边际效用分配
BAMU: Bitstream-Aware Marginal-Utility Allocation for Frozen Pretrained Neural Speech Codecs
arXiv 2608.08432 · 2026-08-11通过二元细化的最优有限区间偏差
Optimal Finite Interval Discrepancy via Binary Refinement
arXiv 2608.08431 · 2026-08-11面向虚拟细胞的人导式因果知识注入
Human-Guided Causal Knowledge Injection for Virtual Cells
arXiv 2608.08430 · 2026-08-11有限资源下的影子层析最优策略
Optimal strategies for shadow tomography with limited resources
arXiv 2608.08429 · 2026-08-11带后继函数的完全n分支序数树的初等理论
The elementary theory of full $n$-branching ordinal trees with successor functions
arXiv 2608.08426 · 2026-08-11PSP：面向 stale 状态与带宽非对称网络的低开销报文级负载均衡
PSP: Low-Overhead Packet-Level Load Balancing for Stale-State and Bandwidth-Asymmetric Networks
arXiv 2608.08425 · 2026-08-11ARC：用于变点定位的增广秩共形化方法——有限样本有效性与分布鲁棒效率
ARC: Augmented-Rank Conformalization for Changepoint Localization --- Finite-Sample Validity and Distribution-Robust Efficiency
arXiv 2608.08424 · 2026-08-11二维Hermite算子的尖锐端点本征函数估计
Sharp Endpoint Eigenfunction Estimates for the Two-Dimensional Hermite Operator
arXiv 2608.08423 · 2026-08-11用于结合文本信息的图像聚类的深度模态共享自表达学习
Learning Deep Modality-Shared Self-Expressiveness for Image Clustering with Textual Information
arXiv 2608.08418 · 2026-08-11面向智能体推荐系统的个性化沟通技能
Personalized Communication Skills for Agentic Recommender Systems
arXiv 2608.08417 · 2026-08-11Tsybakov噪声下的最优学习
Optimal Learning Under Tsybakov Noise
arXiv 2608.08416 · 2026-08-11二维六方磷化硼（h-BP）的热机械特性表征
Thermo-mechanical Characterization of 2D hexagonal Boron Phosphide (h-BP)
arXiv 2608.08415 · 2026-08-11通用可学习概念类的约束学习
Constrained Learning with Universally Learnable Concept Classes
arXiv 2608.08414 · 2026-08-11Tangent：基于大语言模型的智能体应用测试实践的实证研究
Tangent: An Empirical Study of Testing Practices for LLM-Based Agent Applications
arXiv 2608.08413 · 2026-08-11用于扩散给药类器官阵列中表观异质性的仅转运零模型
A transport-only null model for apparent heterogeneity in diffusively dosed organoid arrays
arXiv 2608.08411 · 2026-08-11基于深度学习管道的TGSS ADR1巨型射电源候选体目录
A Catalog of Giant Radio Source Candidates from TGSS ADR1 Using a Deep Learning--based Pipeline
arXiv 2608.08410 · 2026-08-11关于Aizenman--Lieb--Seiringer--Solovej--Yngvason的光晶格硬核玻色气体的注记
Notes on the Optical-Lattice Hard-Core Bose Gas of Aizenman--Lieb--Seiringer--Solovej--Yngvason
arXiv 2608.08409 · 2026-08-11被抽象化：抵制AI研究社区中的异化与无根基抽象
Abstracted Away: Resisting Alienation and Ungrounded Abstraction in AI Research Communities
arXiv 2608.08408 · 2026-08-11立足点是否让竞标者更大胆？多轮收购拍卖中的抢先与多重性
Does a Toehold Make a Bidder Bolder? Preemption and Multiplicity in Multi-Round Takeover Auctions
arXiv 2608.08407 · 2026-08-11重新思考基于学习的影响力最大化：简单神经代理与原生离散搜索
Rethinking Learning-Based Influence Maximization: Simple Neural Surrogates and Native Discrete Search
arXiv 2608.08406 · 2026-08-11鲁棒性还是拥挤：交易策略容量的实验设计
Robustness or Crowding: Experimental Design for Trading Strategy Capacity
arXiv 2608.08405 · 2026-08-11字典序泛函演算及其在泛函演算演算中的应用
Lexicographic functional calculus and its application to functional calculus calculus
arXiv 2608.08404 · 2026-08-11克利福德层级中的深孔
Deep Holes in the Clifford Hierarchy
arXiv 2608.08403 · 2026-08-11智能体驱动的柔性光纤束内窥镜用于体内高分辨率近红外二区（NIR-II）荧光成像
Agentic AI-powered flexible fiber-bundle endoscopy for high-resolution NIR-II fluorescence imaging in vivo
arXiv 2608.08402 · 2026-08-11
