2026-08论文中文摘要
长期测量：迈向对人机交互的纵向理解
Long-term Measurements: Towards a Longitudinal Understanding of Human-AI Interactions
arXiv 2608.02491 · 2026-08-06具有优先方向的各向同性宇宙
Isotropic universes with a preferred direction
arXiv 2608.02458 · 2026-08-06实分析的函数模型
Functionary Models of Real Analysis
arXiv 2608.02423 · 2026-08-06GROVE：基于流视频体验的时间分层记忆的生长与推理
GROVE: Growing and Reasoning over Temporally Stratified Memory from Streaming Video Experience
arXiv 2608.02392 · 2026-08-06揭示并降低金刚石中择优取向氮空位（NV）中心由生长诱导的界面无序
Revealing and reducing growth-induced interfacial disorder in preferentially aligned nitrogen-vacancy centers in diamond
arXiv 2608.02350 · 2026-08-06带分层记忆的Mamba：解决长序列建模中的表示瓶颈
Mamba with Hierarchical Memory: Solving Representation Bottleneck in Long Sequence Modeling
arXiv 2608.02347 · 2026-08-06ECLAIR：一种用于实证软件工程科学发现的因果基础AI框架
ECLAIR: A Causally-Grounded AI Framework for Scientific Discovery in Empirical Software Engineering
arXiv 2608.02323 · 2026-08-06重访拉斯卡尔的频率图分析
Laskar's frequency map analysis revisited
arXiv 2608.02182 · 2026-08-06PhyCheck：面向视频大语言模型的物理规律理解细粒度证据驱动数据集
PhyCheck: Fine-Grained Evidence-Grounded Dataset for Physical Law Understanding in Video-LLMs
arXiv 2608.02150 · 2026-08-06HiResNets：使用中央凹残差流的原生全高清视频识别
HiResNets: Native Full-HD Video Recognition with Foveal Residual Streams
arXiv 2608.02140 · 2026-08-06基于非对称强化学习与自蒸馏的指令条件探索
Instruction-Conditioned Exploration for Reinforcement Learning with Self-Distillation to an Unconditioned Policy
arXiv 2608.02087 · 2026-08-06STEAM：用于EEG解码的分层预训练时空对齐混合专家模型
STEAM: A Spatio-TEmporal Alignment Mixture-of-Experts Model with Hierarchical Pre-training for EEG Decoding
arXiv 2608.02070 · 2026-08-06CompanionBench：一个基于理论、扎根于真实世界的AI情感陪伴基准
CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship
arXiv 2608.02046 · 2026-08-06滑模观测器与浸入不变性观测器的公平比较
A Fair Comparison of Sliding-Mode and Immersion-and-Invariance Observers
arXiv 2608.02000 · 2026-08-06蒸馏前先前瞻：智能体策略内蒸馏中教师指导的未来轨迹验证
Look Ahead Before You Distill: Future Trajectory Validation of Teacher Guidance for Agentic On-Policy Distillation
arXiv 2608.01953 · 2026-08-06尾对尾相互作用诱导的高后牛顿阶动力学：非测地项
High-Post-Newtonian-Order Dynamics Induced by Tail-of-Tail Interactions: The Non-Geodesic Terms
arXiv 2608.01774 · 2026-08-06MemSIF：从结构化交互到面向大语言模型智能体的双轨事实记忆
MemSIF: From Structured Interactions to Dual-Track Fact Memory for LLM Agents
arXiv 2608.01742 · 2026-08-06StreamSplat：流式前馈三维高斯溅射
StreamSplat: Streaming Feed-Forward 3D Gaussian Splatting
arXiv 2608.01659 · 2026-08-06乘积Wasserstein流形上学习的统计力学
Statistical Mechanics of Learning on Product Wasserstein Manifolds
arXiv 2608.01434 · 2026-08-06GenTrack：面向机器人原生运动生成与零样本人形机器人跟踪的物理对齐框架
GenTrack: Physical Alignment for Robot-Native Motion Generation and Zero-Shot Humanoid Tracking
arXiv 2608.01410 · 2026-08-06是解码格式，而非扰动：审计视觉语言模型测试时扩展的基于一致性的选择
It's the Decoding Format, Not the Perturbation: Auditing Consistency-Based Selection for Vision-Language Test-Time Scaling
arXiv 2608.01207 · 2026-08-06关于Datar-Mete-Song极小斜率猜想
On the Datar-Mete-Song minimal slope conjecture
arXiv 2608.01198 · 2026-08-06UniHEAR：面向基于知识的视觉问答的统一异构源注意力检索
UniHEAR: Unified Heterogeneous-Source Attentive Retrieval for Knowledge-Based Visual Question Answering
arXiv 2608.01147 · 2026-08-06VeraRAN：面向异步多接口无线接入网（RAN）规划的预执行验证与事件因果同步修复
VeraRAN: Pre-Actuation Certification and Event-Causal Synchronization Repair for Asynchronous Multi-Interface RAN Plans
arXiv 2608.01047 · 2026-08-06注册中心描述的过期时间存在不均：对模型上下文协议（MCP）漂移的89天测量，以及为何按漂移排名的重新审计无法充分覆盖该问题
Registry Descriptions Go Stale Unevenly: An 89-Day Measurement of Model Context Protocol Drift, and Why Drift-Ranked Re-Auditing Under-Covers It
arXiv 2608.00997 · 2026-08-06暖暴胀中由标量子驱动的暗物质通过紫外冻结产生机制
Scalaron-driven Dark Matter during Warm Inflation via UV Freeze--in
arXiv 2608.00874 · 2026-08-06Freemium 就够了
Freemium Is All You Need
arXiv 2608.00823 · 2026-08-06带有私人位置偏好的顺序拍卖
Order Auctions with Private Position Preferences
arXiv 2608.00786 · 2026-08-06随机多重集合排列中不动点数量的泊松近似
Poisson approximations of the number of fixed points in random multiset permutations
arXiv 2608.00710 · 2026-08-06从受体到全脑动力学的机制桥梁：平均场约化、有效性域与计算权衡
Mechanistic bridges from receptors to whole-brain dynamics: mean-field reductions, validity domains, and computational trade-offs
arXiv 2608.00306 · 2026-08-06ReVoicer：以人为中心、由大语言模型辅助的同行评审对话式语音标注系统
ReVoicer: Conversational Voice Annotation for Human-Centered, LLM-Assisted Peer Review
arXiv 2608.00299 · 2026-08-06截断微分在逆势平均场博弈（MFGs）中的应用
Truncated Differentiation Through Primal-Dual Solvers for Inverse Potential Mean-Field Games
arXiv 2608.00217 · 2026-08-06ExtractBench：模式引导的企业文档抽取基准
ExtractBench: A Benchmark for Schema-Guided Enterprise Document Extraction
arXiv 2607.29677 · 2026-08-06从AdS₃到CFT₂的全息映射
A Holographic Map from AdS$_3$ to CFT$_2$
arXiv 2607.29635 · 2026-08-06相位分辨太赫兹光谱中朗道极化激元的延迟形成
Delayed Formation of Landau Polaritons in Phase-Resolved THz Spectroscopy
arXiv 2607.29501 · 2026-08-06用于一致多参考图像编辑的评估-验证奖励
Evaluation-Verification Reward for Consistent Multi-Reference Image Editing
arXiv 2607.29025 · 2026-08-06DragonCrawl：一种用于可扩展移动端到端测试的生成式意图框架
DragonCrawl: A Generative, Intent-Based Framework for Scalable Mobile End-to-End Testing
arXiv 2607.28750 · 2026-08-06包含核效应和盖根鲍尔多项式部分子分布的$xF_3$的拉普拉斯空间分析
Laplace-Space Analysis of $xF_3$ Including Nuclear Effects and Gegenbauer-Polynomial Parton Distributions
arXiv 2607.28105 · 2026-08-06梯度Kähler-Ricci孤立子上多重调和函数的$L^p$刘维尔定理
$L^p$ Liouville theorems for pluriharmonic functions on gradient Kähler-Ricci solitons
arXiv 2607.28057 · 2026-08-06ChronoMem：大语言模型智能体记忆的版本控制与语义回滚
ChronoMem: Version Control and Semantic Rollback for Large Language Model Agent Memory
arXiv 2607.27773 · 2026-08-06DESI DR2结果IV：来自莱曼α森林的Alcock-Paczyński测量及宇宙学约束
DESI DR2 Results IV: Alcock-Paczyński Measurements from the Lyman Alpha Forest and Cosmological Constraints
arXiv 2607.27410 · 2026-08-06开发者使用本地大语言模型生成和评估的代码导航体验陌生代码库调试的研究
How Developers Experience Debugging Unfamiliar Codebases with Code Tours Generated and Evaluated by Local LLMs
arXiv 2607.26987 · 2026-08-06李-杨零点与粒子涨落
Lee-Yang Zeros And Particle Fluctuations
arXiv 2607.26975 · 2026-08-06带不精确梯度的凸(L0, L1)-光滑优化的归一化一阶方法
Normalized First-Order Methods for Convex (L0, L1)-Smooth Optimization with Inexact Gradients
arXiv 2607.26969 · 2026-08-06欧几里得球上诺伊曼拉普拉斯算子的波利亚猜想
Pólya's Conjecture for the Neumann Eigenvalues on Euclidean Balls
arXiv 2607.25958 · 2026-08-06护盾斯特拉尔
Shieldstral
arXiv 2607.25857 · 2026-08-06CORF-GS：通过耦合光学-射频高斯格点法进行实时无线辐射场重建
CORF-GS: Real-Time Wireless Radiance Field Reconstruction via Coupled Optical-RF Gaussian Splatting
arXiv 2607.25569 · 2026-08-06从分析到参数化：通过智能手机加速度计实现物理引导的声学窃听
From Profiling to Parameterization: Physics-Guided Acoustic Eavesdropping via Smartphone Accelerometers
arXiv 2607.25461 · 2026-08-06打破周期性假设：通过图谱低秩学习实现鲁棒张量多视图聚类
Breaking the Periodicity Assumption: Robust Tensorial Multi-View Clustering via Graph-Spectral Low-Rank Learning
arXiv 2607.25295 · 2026-08-06MyoCardBench：用于评估临床真实心血管护理场景中大型语言模型的真实世界数据基准
CardioBench: A Real-World Data Benchmark for Evaluating Large Language Models in Clinically Authentic Cardiovascular Care Scenarios
arXiv 2607.25186 · 2026-08-06
