2026-08论文中文摘要
CultureVidBench：文本到视频生成中的文化理解基准测试
CultureVidBench: Benchmarking Cultural Understanding in Text-to-Video Generation
arXiv 2608.01942 · 2026-08-28不同次数多项式的两步幂等遍历平均的结构
Structure of 2-step nilpotent ergodic averages for distinct-degree polynomials
arXiv 2607.29368 · 2026-08-28与离散时间量子行走相关的相互作用Fock空间上的Poisson算子
Poisson operator on the interacting Fock space associated with a discrete-time quantum walk
arXiv 2607.29345 · 2026-08-28由对数拉普拉斯驱动的半线性热方程的尖锐寿命二分性与阈值现象
Sharp Lifespan Dichotomies and Threshold Phenomena for Semilinear Heat Equations Driven by the Logarithmic Laplacian
arXiv 2607.29318 · 2026-08-28扭曲单层-菱面石墨烯中的新兴网络模型
Emerging network model in a twisted monolayer-rhombohedral graphene
arXiv 2607.29049 · 2026-08-28使用大型张量网络对Lipkin-Meshkov-Glick模型的量子模拟进行基准测试
Benchmarking Quantum Simulations of the Lipkin-Meshkov-Glick Model Using Large Tensor Networks
arXiv 2607.28570 · 2026-08-28用AI生成的智能手机物理实验室研究圆周运动
Uniform and Accelerated Circular Motion with a Smartphone: A No-Code, AI-Generated Browser Laboratory
arXiv 2607.28352 · 2026-08-28大型视觉语言模型（LVLMs）能否揭示视觉错觉背后的真相？感知与推理能力分析
Can LVLMs Uncover the Truth Behind Visual Illusions? An Analysis of Perceptual and Reasoning Capabilities
arXiv 2607.27747 · 2026-08-28由逆平均曲率流生成的压缩映射
Contracting Transport Maps on Riemannian Manifolds
arXiv 2607.27711 · 2026-08-28算术等价数域的高阶代数K-群
On the higher algebraic $K$-groups of arithmetically equivalent number fields
arXiv 2607.26685 · 2026-08-28图像分类神经网络中高权重神经元是否重要？
Are the High-weight Neurons the Important Ones in Image Classification Neural Networks?
arXiv 2607.25529 · 2026-08-28量子检测\(t\bar{t}\)系统中的CP破坏：断层扫描
Quantum detection of CP violation in the $t\bar{t}$ system: tomography
arXiv 2607.25034 · 2026-08-28量子检测$t\bar{t}$系统中的CP破坏：产生
Quantum detection of CP violation in the $t\bar{t}$ system: production
arXiv 2607.25029 · 2026-08-28一种受模型预测控制启发的量子算法
A Model Predictive Control-Inspired Quantum Algorithm
arXiv 2607.24992 · 2026-08-28基于大语言模型的自动出价的策略感知参数高效适配
Strategy-Aware Parameter-Efficient Adaptation for LLM-based Auto-Bidding
arXiv 2607.24232 · 2026-08-28TriShieldRAG：一种针对检索增强生成中知识腐败的三环深度防御框架
TriShieldRAG: 3 Rings, One Blind Spot in Layered Defenses for Retrieval-Augmented Generation
arXiv 2607.23838 · 2026-08-28ATLAS：用于在一小时内实现高效同态推理的变压器自动近似
ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour
arXiv 2607.23478 · 2026-08-28变分自由能分解的一个唯一性定理
A Uniqueness Theorem for Exact Separable Log-Partition Decompositions
arXiv 2607.22710 · 2026-08-28群的图积的同构与自同构
Isomorphisms and automorphisms of graph products of groups
arXiv 2607.22303 · 2026-08-28特征函数、自由边界与时频局部化
Eigenfunctions, free boundaries, and time-frequency localization
arXiv 2607.21590 · 2026-08-28具有非稠密像的实多项式凯勒映射的一般次数
Generic degrees of real polynomial Keller maps with non-dense image
arXiv 2607.21572 · 2026-08-28利用多频时域分量分离减轻宇宙微波背景观测中的大气干扰
Atmosphere mitigation in CMB observations using multi-frequency time-domain component separation
arXiv 2607.21432 · 2026-08-28PercepCap：具有结构化时空感知的视频字幕生成器
PercepCap: Video Captioner with Structured Spatio-Temporal Perception
arXiv 2607.20389 · 2026-08-28通过冲突感知不相交参数训练实现统一预测与规划
Unified Prediction and Planning via Conflict-Aware Disjoint Parameter Training
arXiv 2607.19971 · 2026-08-28无需经过非手性状态的波手性反转
Wave chirality reversal without passing achirality
arXiv 2607.18822 · 2026-08-28使用编码智能体进行自动研究：古兰经诵读数据上的泛化器和指标最大化器
Autoresearch with Coding Agents: Generalizers and Metric-Maximizers on Quran Recitation Data
arXiv 2607.18064 · 2026-08-28国家层面与领域层面：调查重新设计下分层贝叶斯可信区间的覆盖特性
National Versus Domain: Coverage Properties of HB Credible Intervals Under Survey Redesign
arXiv 2607.17229 · 2026-08-28解决ARC-AGI-3编码智能体是否需要可执行世界模型、简化和验证？
Do Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3?
arXiv 2607.15439 · 2026-08-28轴对称通道中电解质溶液的扩散渗透
Diffusioosmosis of electrolyte solutions in axisymmetric channels
arXiv 2607.15199 · 2026-08-28克利福德半群与幺半格罗滕迪克构造
Clifford semigroups and the monoidal Grothendieck construction
arXiv 2607.12944 · 2026-08-28用于湍流减阻的闭环壁面控制器的无梯度学习
Gradient-free learning of a closed-loop wall controller for turbulent drag reduction
arXiv 2607.12626 · 2026-08-28无元数操作的非二元自底向上成分句法分析
Non-binary bottom-up constituency parsing without arity actions
arXiv 2607.10591 · 2026-08-28在域转移下对用于乳腺钼靶成像的基础模型的稳健性进行基准测试
Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
arXiv 2607.10358 · 2026-08-28关于\(\mathbb{L}(\mathbb{X}, \mathbb{Y})\)的球覆盖性质
On the Ball Covering Property of $\mathbb{L}(\mathbb{X}, \mathbb{Y})$
arXiv 2607.10353 · 2026-08-28无声冻结：预测低精度训练何时停止学习
Auditing Invisible Weight Updates with Reference Traces
arXiv 2607.09800 · 2026-08-28穿越Rindler防火墙的非线性粒子探测器
Nonlinear particle detectors across the Rindler firewall
arXiv 2607.09660 · 2026-08-28通过读出电流非线性实现微波动态电感探测器的可调响应度和带宽
Tunable responsivity and bandwidth in microwave kinetic inductance detectors via readout current nonlinearity
arXiv 2607.09178 · 2026-08-28追溯宇宙正午以来不同环境下黑洞和星系的增长
Tracing black hole and galaxy growth across environments since cosmic noon
arXiv 2607.07793 · 2026-08-28分段多面体映射的极大单调性
Maximal monotonicity of piecewise polyhedral mappings
arXiv 2607.07358 · 2026-08-28非最小k完美哈希：紧密下界及在快速静态哈希表中的应用
Non-Minimal $k$-Perfect Hashing: Tight Lower Bounds and an Application to Fast Static Hash Tables
arXiv 2607.07257 · 2026-08-28多教师在线策略蒸馏中的行为杠杆不平衡
When Top-K Misses the Decision: Tool-Call Drift in Multi-Teacher On-Policy Distillation
arXiv 2607.07050 · 2026-08-28赫瓦尔53年UBV测光的精确归约与标准化
Refined reduction and standardisation of 53 years of UBV photometry at Hvar
arXiv 2607.06104 · 2026-08-28基于同家族架构指导的大语言模型驱动神经网络生成：区分迁移与适应
Curating Same-Family Neural Networks for LLM-Guided Model Improvement: A Controlled Case Study
arXiv 2607.05704 · 2026-08-28具有多个传感器和一个控制器的网络控制的下界及其在跟踪高斯 - 马尔可夫源中的应用
Lower Bound of Networked Control with Multiple Sensors and One Controller And The Application to Tracking Gaussian-Markov Source
arXiv 2607.04172 · 2026-08-28模椭圆曲线与双曲均匀化
Modular elliptic curves and hyperbolic uniformization
arXiv 2607.03830 · 2026-08-28一维Pólya-Szegő原理的一个变体
A variation on the Pólya-Segő principle in one dimension
arXiv 2607.03450 · 2026-08-28PPE-Bench：用于评估公私纠缠下MLLM遗忘能力的基准测试
PPE-Bench: A Benchmark for Evaluating MLLM Unlearning under Private-Public Entanglement
arXiv 2607.02897 · 2026-08-28NEvo: 神经引导的动态视觉选择性视频合成
NEvo: Neural-Guided Evolutionary Video Synthesis for Dynamic Visual Selectivity
arXiv 2607.02317 · 2026-08-28基于矩阵分解MDP的候选生成规划
Planning over Matrix-Factorization MDPs for Candidate Generation
arXiv 2607.02115 · 2026-08-28用小碰撞和大碰撞系统中的重夸克偶素探测热QCD介质
Probing hot QCD medium with heavy quarkonium in small and large collision systems
arXiv 2607.02042 · 2026-08-28
