2026-08论文中文摘要
ExploraTwin:一个用于数字孪生模拟的非营利研究平台
ExploraTwin, a Non-Profit Research Platform for Digital Twin Simulations
arXiv 2608.20539 · 2026-08-26Metag:用于构建智能体元评审能力的数据集
Metag: A dataset to build agentic meta-reviewing capabilities
arXiv 2608.20488 · 2026-08-26基于最优传输的视觉信息聚合用于视频语言模型的令牌压缩
Aggregating Visual Information with Optimal Transport for VideoLM Token Compression
arXiv 2608.20473 · 2026-08-26ARGUS:基于心理理论(Theory-of-Mind)的论证生成,结合策略感知规划与知识接地
ARGUS: Theory-of-Mind Guided Argument Generation with Strategy-Aware Planning and Knowledge Grounding
arXiv 2608.20405 · 2026-08-26若它像套利:具有可判定结构等价性的协议无关检测
If It Walks Like an Arbitrage: Protocol-Agnostic Detection with Decidable Structural Equivalence
arXiv 2608.20377 · 2026-08-26用于评估大型语言模型在临床登记处抽象任务中性能的歧义分类:一项多中心前瞻性研究
An ambiguity taxonomy for evaluating large language model performance on clinical registry abstraction: a multi-site prospective study
arXiv 2608.20373 · 2026-08-26关于Bray猜想的证明
On the proof of Bray's conjecture
arXiv 2608.20215 · 2026-08-26你看不见的是你所学的:受限证据可见性有利于共享基因组语言模型社群中的组合泛化
What You Can't See Is What You Learn: Slot-Selective Evidence Masking Favors Compositional Generalization in Shared-Genome Language-Model Societies
arXiv 2608.20054 · 2026-08-26ExPhy:多目标轨迹预测中显式物理属性学习的基准
ExPhy: A Benchmark for Explicit Physical Property Learning in Multi-Object Trajectory Forecasting
arXiv 2608.20009 · 2026-08-26色多项式与列表着色函数相等性的非持久性
Non-persistence of equality between chromatic polynomials and list-color functions
arXiv 2608.19773 · 2026-08-26无真实值的信用:针对执行重放的大语言模型智能体的步骤级信用分配审计
Credit Without Ground Truth: Auditing Step-Level Credit Assignment in LLM Agents Against Executed Replay
arXiv 2608.19760 · 2026-08-26双星系统中中子星的初始 kick 速度
The Kick Velocities of Neutron Stars in Binary Systems
arXiv 2608.19690 · 2026-08-26视频捕捉的同震地表断层位移:三次地震中的图像跟踪
Co-seismic surface fault displacement captured in videos: image tracking across three earthquakes
arXiv 2608.19656 · 2026-08-26Block3D:基于分块扩散的高效文本到3D生成
Block3D: Efficient Text-to-3D Generation via Block-Wise Diffusion
arXiv 2608.19567 · 2026-08-26委托还是自行操作?理解混合人机界面中的用户行为
Delegating or Doing? Understanding User Behavior in Hybrid Human-Agent Interfaces
arXiv 2608.19551 · 2026-08-26分数量子霍尔-超导体异质结构中零模的干涉特征
Interferometric Signatures of Zero Modes in Fractional Quantum Hall-Superconductor Heterostructures
arXiv 2608.19312 · 2026-08-26与损耗腔耦合的SYK模型中耗散增强的 scrambling 效应
Dissipation-enhanced scrambling in the SYK model coupled to a lossy cavity
arXiv 2608.19310 · 2026-08-26Hessian商方程的内部Hessian估计
Interior Hessian estimates for Hessian quotient equations
arXiv 2608.19087 · 2026-08-26TractorBeam:通过协作机器标注实现个性化AI意义建构支持
TractorBeam: Personalized AI Sensemaking Support via Collaborative Machine Annotation
arXiv 2608.18994 · 2026-08-26CauSec:揭示静态漏洞分析性能的因果驱动因素
CauSec: Unboxing the Causal Drivers of Static Vulnerability Analysis Performance
arXiv 2608.18876 · 2026-08-26GEAR:面向表格基础模型两阶段蒸馏的生成式扩展与真实锚定
GEAR: Generative Expansion and Real Anchoring for Two-Stage Distillation of Tabular Foundation Models
arXiv 2608.18849 · 2026-08-26K(10,3)的二元电压覆盖:上同调、对称轨道与局部K(7,3)图
Type-Voltage Covers and Finite Locally Kneser Graphs
arXiv 2608.18754 · 2026-08-26在曲线上寻找适当的一般点及其在刚性实闭域构造中的应用
Finding suitably generic points on curves with an application to the construction of rigid real closed fields
arXiv 2608.18652 · 2026-08-26环面上弹性Neumann-Poincaré算子的傅里叶块的本质谱与本征值渐近行为
Essential spectra and eigenvalue asymptotics of Fourier blocks of the elastic Neumann-Poincaré operator on tori
arXiv 2608.18499 · 2026-08-26罗马诺夫三元组逻辑的形式化验证:一种针对滑动窗口3-CNF的已验证过滤器及在结构化公式中的应用
Formal Verification of Romanov's Triplet Logic: A Verified Filter for Sliding-window 3-CNF with Application to Structured Formulas
arXiv 2608.18445 · 2026-08-26基于归一化带符号原积分与显式参考策略的任意时间求解器评估——扩展版
Anytime Solver Evaluation with a Normalized Signed Primal Integral and Explicit Reference Policies - Extended Version
arXiv 2608.18288 · 2026-08-26从语料到协同进化的能力:面向通用图像生成的以能力为中心的数据设计
From Corpora to Co-Evolving Capabilities: Capability-Centric Data Design for Generalist Image Generation
arXiv 2608.18076 · 2026-08-26CoinVE-200K:用于组合式指令引导视频编辑的大规模高质量数据集
CoinVE-200K: A Large-Scale High-Quality Dataset for Compositional Instruction-Guided Video Editing
arXiv 2608.17566 · 2026-08-26一对显式信道对的正阶Rényi可加性违背的严格认证
Exact certification of a positive-order Rényi additivity violation for an explicit channel pair
arXiv 2608.17376 · 2026-08-26非共线反铁磁体Mn₃Sn中的风车自旋动力学及其诱导的反常霍尔效应
Windmill Spin Dynamics and Its Induced Anomalous Hall Effect in Noncollinear Antiferromagnet Mn3Sn
arXiv 2608.17225 · 2026-08-26二维拓扑量子码最小重量译码的近似难度
Hardness of approximation for minimum-weight decoding of two-dimensional topological quantum codes
arXiv 2608.17109 · 2026-08-26双重否定何时是Scott连续的?
When is double negation Scott continuous?
arXiv 2608.17107 · 2026-08-26TomoSphero:用于球面网格上行星与太阳层析成像的快速可微投影器
TomoSphero: Fast Differentiable Projector for Planetary and Solar Tomography on Spherical Grids
arXiv 2608.16960 · 2026-08-26重构:从预发表参考文献中恢复研究思路的盲基准
Reconstruction: A Blind Benchmark for Recovering Research Ideas from Pre-Publication Bibliographies
arXiv 2608.16645 · 2026-08-26表征还不够:用于阿片类药物使用障碍非接触式压力与渴求感感知的身体局部热证据
Representation Is Not Enough: Body-Localized Thermal Evidence for Contactless Stress and Craving Sensing in Opioid Use Disorder
arXiv 2608.16087 · 2026-08-26TRACER:平衡大语言模型增强型持续推荐的稳定性-可塑性-认知性三难困境
TRACER: Balancing Stability-Plasticity-Cognitivity Trilemma for LLM Enhanced Continual Recommendation
arXiv 2608.16075 · 2026-08-26GOD:通过深度嫁接提升序列推荐的泛化能力
GOD: Enhancing Generalization via Deep Grafting for Sequential Recommendation
arXiv 2608.16073 · 2026-08-26无预设相位分布的彩色超透镜的直接相机-信息优化
Information-optimized color metalenses for camera imaging
arXiv 2608.15907 · 2026-08-26深度非弹性散射(DIS)全局事件形状中强子化效应的普适性与运动学依赖关系
Universality and Kinematic Dependence of Hadronization Effects in DIS Global Event Shapes
arXiv 2608.15898 · 2026-08-26双椭圆亏格二轨迹上平坦度量的谱行列式
Spectral determinants of flat metrics on the bielliptic genus-two locus
arXiv 2608.15752 · 2026-08-26埃尔米特流形上复赫斯方程的正则化与赫尔德连续性
Regularization and Hölder continuity for complex Hessian equations on Hermitian manifolds
arXiv 2608.15552 · 2026-08-26MAPLE:MoE自适应即插即用分层专家分配
MAPLE: MoE Adaptive Plug-and-play Layer-wise Expert allocation
arXiv 2608.15299 · 2026-08-26基于离散近似的复杂系统结构保持可视化:在Argo数据中的应用
Structure-Preserving Visualization of Complex Systems through Discrete Approximation: An Application to Argo Data
arXiv 2608.15013 · 2026-08-26智能体在自动研究(AutoResearch)中如何失败?针对100项真实前沿研究任务的端到端诊断评估
How Do Agents Fail on AutoResearch: End-to-End Diagnostic Evaluation on 100 Real-World Frontier Research Tasks
arXiv 2608.14905 · 2026-08-26高阶非绝热性决定声子谱的温度依赖关系
Higher-order nonadiabaticity governs the temperature dependence of the phonon spectrum
arXiv 2608.14882 · 2026-08-26ER-KANs:面向数据稀缺的科学机器学习的高效鲁棒柯尔莫哥洛夫-阿诺德网络
ER-KANs: Efficient and Robust Kolmogorov-Arnold Networks for Data-Scarce Scientific Machine Learning
arXiv 2608.14773 · 2026-08-26沿等分布螺旋的定量时间平均球面均值:丢番图速率与最优性
Quantitative time-averaged spherical means along equidistributed spirals: Diophantine rates and limits of uniformity
arXiv 2608.14607 · 2026-08-26代数曲面上点的近对角加性能量界
Near diagonal additive energy bound for points on algebraic surfaces
arXiv 2608.14467 · 2026-08-26正数量曲率与体积增长
Positive Scalar Curvature and Volume Growth
arXiv 2608.14438 · 2026-08-26智能爆炸的动力学
The Dynamics of Intelligence Explosions
arXiv 2608.14426 · 2026-08-26