2026-08论文中文摘要
体育领域的大模型综述
A Survey of Large Models in Sports
arXiv 2608.14377 · 2026-08-17CoRun：填充对于确定性大语言模型（LLM）推理而言简单且高效
CoRun: Padding is Simple and Efficient for Deterministic LLM Inference
arXiv 2608.14376 · 2026-08-17错误但有用：多智能体消息中超越答案正确性的轨迹价值
Wrong but Useful: Trajectory Value Beyond Answer Correctness in Multi-Agent Messages
arXiv 2608.14375 · 2026-08-17热带气旋热结构的演变特征：热带气旋热相空间
Characterizing the Evolution of Tropical Cyclone Thermal Structure: A Tropical Cyclone Thermal Phase Space
arXiv 2608.14374 · 2026-08-17识别冒名者：通过跨实体特征置换学习物理一致性的自监督学习
Catching the Imposter: Self-Supervised Learning of Physical Coherence with Cross-Entity Feature Permutations
arXiv 2608.14372 · 2026-08-17Hurwitz类数关系与 mock 模形式
Hurwitz class number relations and mock modular forms
arXiv 2608.14371 · 2026-08-17一种基于混合大语言模型（LLM）的业务流程模型自动安全标注生成框架
A Hybrid LLM-Based Framework for Automated Security Annotation Generation in Business Process Models
arXiv 2608.14370 · 2026-08-17揭示浅陷阱、深陷阱与俄歇陷阱在三阳离子钙钛矿中电荷载流子复合的作用
Unraveling the Roles of Shallow, Deep and Auger Trapping in Charge Carrier Recombination in Triple-Cation Perovskites
arXiv 2608.14368 · 2026-08-17注意长尾：理解业务流程中延迟检测的难度
Mind the Long Tail: Understanding the Difficulty of Delay Detection in Business Processes
arXiv 2608.14367 · 2026-08-17Sentinel-1 SAR影像中的弱监督极地低压分割
Weakly Supervised Polar Low Segmentation in Sentinel-1 SAR Imagery
arXiv 2608.14366 · 2026-08-171+1维正弦-戈登模型中场指数的形状因子
Form factors of field exponentials in the 1+1 dimensional Sine-Gordon model
arXiv 2608.14365 · 2026-08-17认知张力：通过纠缠理论重构可视化协同设计
Epistemic Tensions: Reframing A Visualization Co-Design through Entanglement Theory
arXiv 2608.14364 · 2026-08-17钱德拉X射线天文台ACIS污染层2010年至2026年的演化
The Evolution of the ACIS Contamination Layer on the Chandra X-ray Observatory from 2010 to 2026
arXiv 2608.14363 · 2026-08-17通过氦离子辐照调控SrVO₃薄膜的晶体场与轨道Widom线
Tuning crystal-fields by He-irradiation and orbital Widom line in SrVO$_3$ films
arXiv 2608.14362 · 2026-08-17语言模型表示中几何复杂度的局部与全局 regime
Local and Global Regimes of Geometric Complexity in Language Model Representations
arXiv 2608.14361 · 2026-08-17关于弦代数的几何模型：曲面的唯一性与红点的存在性
On Geometric Models of String Algebras: Uniqueness of Surfaces and Existence of Red Punctures
arXiv 2608.14360 · 2026-08-17利用神经架构搜索设计可持续的联邦学习即服务
Designing Sustainable Federated Learning as a Service using Neural Architecture Search
arXiv 2608.14359 · 2026-08-17厄米特曲率流与HKT几何
Hermitian curvature flow and HKT geometry
arXiv 2608.14358 · 2026-08-17单结unimon电路上的原生多量子比特门
Native multi-qubit gates on a single-junction unimon circuit
arXiv 2608.14357 · 2026-08-17设计普惠加密资产争议解决：面向弱势用户的AI与智能合约混合式在线争议解决框架
Designing Inclusive Crypto-Asset Dispute Resolution A Hybrid AI and Smart Contract Online Dispute Resolution Framework for Vulnerable Users
arXiv 2608.14356 · 2026-08-17解耦共享表示可改进形态-转录组整合
Disentangled Shared Representations Improve Morpho-Transcriptomic Integration
arXiv 2608.14355 · 2026-08-17基于量子几何带干涉的外尔半金属中的轨道霍尔效应
Orbital Hall Effect in Weyl Semimetals from quantum geometric band interference
arXiv 2608.14353 · 2026-08-17ATLAS：通过大语言模型引导的抽象与自动机学习发现智能体策略
ATLAS: Discovering Agent Strategies through LLM-Guided Abstraction and Automata Learning
arXiv 2608.14352 · 2026-08-17用于直接丙烯环氧化的金属-氮-碳电催化剂的多维度设计
Multidimensional Design of Metal-Nitrogen-Carbon Electrocatalysts for Direct Propylene Epoxidation
arXiv 2608.14351 · 2026-08-17将Goldberg正合序列推广到图与单纯复形的辫群
Extending Goldberg's Exact Sequence to Braid Groups of Graphs and Simplicial Complexes
arXiv 2608.14350 · 2026-08-17全场与布洛赫周期因子离散化：精度与幻影模式
Full-field and Bloch-periodic-factor discretizations: Accuracy and phantom modes
arXiv 2608.14348 · 2026-08-17全息QCD与耦合道色散关系下的π介子和K介子D项
Pion and kaon D terms from holographic QCD and coupled-channel dispersion relations
arXiv 2608.14347 · 2026-08-17平流扩散壳模型中最优混合的长时间行为
Long-time behavior of optimal mixing in an advection-diffusion shell model
arXiv 2608.14346 · 2026-08-17分子电子跃迁的TDDFRT表示是唯一的吗？
Is the TDDFRT representation of a molecular electronic transition unique?
arXiv 2608.14345 · 2026-08-17高维时间序列中的变点检测与定位
Change Point Detection and Localization in High-Dimensional Time Series
arXiv 2608.14344 · 2026-08-17切片定理的一个确定性证明
A Deterministic Proof of the Slicing Theorem
arXiv 2608.14342 · 2026-08-17三粒子对电弱企鹅B介子衰变的增强贡献
Enhanced Three-Particle Contribution to Electroweak Penguin $B$-Meson Decays
arXiv 2608.14341 · 2026-08-17粘合抛物特征层
Gluing parabolic character sheaves
arXiv 2608.14340 · 2026-08-17一致稠密3-均匀超图中的紧哈密顿圈
Tight Hamiltonian Cycles in Uniformly Dense $3$-Graphs
arXiv 2608.14338 · 2026-08-17环面阿贝尔纤维化的权分解
Weight decomposition for toroidal abelian fibrations
arXiv 2608.14337 · 2026-08-17Wasserstein分布鲁棒优化的收缩路径启发式算法
A Shrinkage Path Heuristic for Wasserstein Distributionally Robust Optimization
arXiv 2608.14336 · 2026-08-17二维能量约束最优冷却问题的边界优化
Optimizing bounds for energy-constrained optimal cooling problems in two dimensions
arXiv 2608.14334 · 2026-08-17超越容量：通过具有直接GPU和HBM路径的高带宽闪存实现可扩展的MoE大语言模型推理
Beyond Capacity: Scalable MoE LLM Inference via High-Bandwidth Flash with Direct GPU and HBM Paths
arXiv 2608.14333 · 2026-08-17CORAL：面向基于激光雷达的目标导向城市驾驶的课程优化奖励适配
CORAL: Curriculum-Optimized Reward Adaptation for LiDAR-Based Goal-Directed Urban Driving
arXiv 2608.14332 · 2026-08-17混合量子态学习的平均情况困难性与密码学的等价性
Equivalence Between Average-Case Hardness of Learning and Cryptography for Mixed Quantum States
arXiv 2608.14331 · 2026-08-17用于形态学到转录组学预测的程序空间扩散模型
Program-space Diffusion for Morphology-to-Transcriptomics Prediction
arXiv 2608.14330 · 2026-08-17面向基于原则的监管的LLM作为评判者的四轴可信度基准
A Four-Axis Trustworthiness Benchmark for LLM-as-Judge in Principle-Based Regulation
arXiv 2608.14329 · 2026-08-17降阶建模及其在逆状态估计中的应用
Reduced Order Modeling and Applications to Inverse State Estimation
arXiv 2608.14328 · 2026-08-17量子快照揭示紧致共形边界模式
Quantum Snapshots Reveal a Compact Conformal Boundary Mode
arXiv 2608.14327 · 2026-08-17带有删失观测的贝叶斯序贯搜索
Bayesian Sequential Search with Censored Observations
arXiv 2608.14326 · 2026-08-17依赖关系感知的稀疏神经网络架构用于股票收益预测
Dependence-Informed Sparse Neural Architecture for Stock Return Prediction
arXiv 2608.14323 · 2026-08-17用反例引导的精化方法求解量化布尔公式（QBF）
Solving QBF with Counterexample Guided Refinement
arXiv 2608.14322 · 2026-08-17TRIAGE：面向标注高效的半监督视网膜OCT分类的风险控制伪标签准入框架
TRIAGE: Risk-Controlled Pseudo-Label Admission for Annotation-Efficient Semi-Supervised Retinal OCT Classification
arXiv 2608.14321 · 2026-08-17AnchorBench：针对大语言模型中锚定效应的多路径基准测试
AnchorBench: A Multi-Pathway Benchmark for the Anchoring Effect in LLMs
arXiv 2608.14320 · 2026-08-17面向触觉互联网约束的QUIC中BBRv2的鲁棒感知约束贝叶斯调优
Robust Constraint-Aware Bayesian Tuning of BBRv2 for QUIC under Tactile Internet Constraints
arXiv 2608.14318 · 2026-08-17
