2026-09论文中文摘要
面向可实时部署在边缘设备的虚拟感知的、具备蒸馏辅助的低延迟激活正则化稀疏神经算子
Low-Latency Activation-Regularized Sparse Neural Operators with Distillation Assistance Towards Real-Time Neuromorphic Virtual Sensing
arXiv 2608.23987 · 2026-09-29Luce：用于3D资产生成的可重光照高斯模型
Luce: Relightable Gaussians for 3D Asset Generation
arXiv 2608.23943 · 2026-09-29ISCSLP 2026 真实场景视听语音增强挑战赛
The ISCSLP 2026 Real-World Audio-Visual Speech Enhancement Challenge: A Benchmark with Natural Mixtures and Degraded Video
arXiv 2608.23759 · 2026-09-29通过速度匹配扩展扩散模型的强化学习
Scaling Reinforcement Learning for Diffusion Models via Velocity Matching
arXiv 2608.23664 · 2026-09-29基于三轴身体震波的物理约束深度学习血压监测模型
Physics-Constrained Deep Learning Model for Contactless Blood Pressure Monitoring from Triaxial Bodyseismography
arXiv 2608.23562 · 2026-09-29在结构化表面上达到毛细上升的热力学极限
Approaching the theoretical limit of wicking on textured surfaces
arXiv 2608.23487 · 2026-09-29绿色专利分类中的系统性偏差：隐性绿色与虚假绿色
Systematic Bias in Green Patent Classification: Silent Green and False Green
arXiv 2608.23420 · 2026-09-29局部几何偏序集的Orlik–Solomon代数与复阿贝尔排列的上同调
The Orlik--Solomon algebra of a locally geometric poset and cohomology of complex abelian arrangements
arXiv 2608.23384 · 2026-09-29时间序列预测中哪些历史数据重要？通过未来监督学习预测相关性
Which Histories Matter for Time Series Forecasting? Learning Predictive Relevance with Future Supervision
arXiv 2608.23221 · 2026-09-29子因子交换平方的结构与动力学性质
Structural and Dynamical Properties of Subfactor Commuting Squares
arXiv 2608.23125 · 2026-09-29神经玻尔兹曼方程
Neural Boltzmann Equations
arXiv 2608.23022 · 2026-09-29掩码并非模型：审计注意力、状态空间与混合序列模型中的前缀不变性
The Mask Is Not the Model: Auditing Prefix Invariance in Attention, State-Space, and Hybrid Sequence Models
arXiv 2608.22876 · 2026-09-29用于autoPETV大赛的三阶段涂鸦自适应课程学习方法
Three-Phase Scribble-Adaptive Curriculum Learning for autoPETV Grand Challenge
arXiv 2608.22096 · 2026-09-29SkillBloat：通过大语言模型编码智能体中的技能注入实现的令牌放大攻击
SkillBloat: Token Amplification Attacks via Skill Injection in LLM Coding Agents
arXiv 2608.21929 · 2026-09-29训练需要可信的世界：用于智能体学习的经过验证的合成Web环境
Training Needs Trustworthy Worlds: Verified Synthetic Web Environments for Agent Learning
arXiv 2608.21898 · 2026-09-29从求解器反馈到可信规划：用于符号规划的多角色强化学习
From Solver Feedback to Faithful Plans: Multi-Role Reinforcement Learning for Symbolic Planning
arXiv 2608.21897 · 2026-09-29弱监督视频异常检测中的帧级评估大多衡量的是视频级排名
Look Inside Each Video: Rethinking How Video Anomaly Detection Is Evaluated
arXiv 2608.21854 · 2026-09-29LitReview Arena：基于对战式同行评审平台的文献综述智能体评估
LitReview Arena: Evaluating Literature Review Agents with Battle-Style Peer Review Platform
arXiv 2608.21374 · 2026-09-29银河系的随机磁场
The Random Magnetic Field of the Milky Way
arXiv 2608.21293 · 2026-09-29VLM 的回答并非异常分数：无训练视频异常检测中的排名压缩
A VLM Answer Is Not an Anomaly Score: Rank Compression Across Image and Video Anomaly Detection
arXiv 2608.21244 · 2026-09-29评估大型语言模型在国际海运危险货物规则合规性上的性能
Evaluating Large Language Model Performance on International Maritime Dangerous Goods Code Compliance
arXiv 2608.21036 · 2026-09-29无需解决，仅需比较：用于LLM智能体运行时干预的微型顾问
Don't Solve, Just Compare: Tiny Advisors for Runtime Intervention in LLM Agents
arXiv 2608.21027 · 2026-09-29两栖动物从水生到陆生的过渡揭示了嗅觉的物理极限
Amphibian water to land transition reveals physical limits of olfaction
arXiv 2608.20765 · 2026-09-29n-2个反射对称下的对数Brunn-Minkowski不等式
Logarithmic Brunn--Minkowski Inequality under $n-2$ Reflection Symmetries
arXiv 2608.20730 · 2026-09-29基于F₂线性代数学习精确的NVIDIA SASS编码器
Learning Exact NVIDIA SASS Encoders with $\mathbb{F}_2$ Linear Algebra
arXiv 2608.20532 · 2026-09-29关于Verlinde范畴$\boldsymbol{\text{Ver}}_p$的Drinfeld中心
On the Drinfeld center of the Verlinde category $\Ver_p$
arXiv 2608.20465 · 2026-09-29中子星中的迹异常与有效拓扑源
Trace Anomaly and Effective Topological Sources in Neutron Stars
arXiv 2608.20460 · 2026-09-29Bernard-Letac公平采样构造的算法、复杂性与熵
Algorithms, Complexity, and Entropy of the Bernard-Letac Fair-Sampling Construction
arXiv 2608.20234 · 2026-09-29具备可证明双准则保证的大语言模型智能体最优技能选择
Optimal Skill Selection for LLM Agents with Provable Bicriteria Guarantees
arXiv 2608.19993 · 2026-09-29学习如何遗忘：面向长上下文稀疏注意力的微调
Learning how to Forget: Fine-tuning for Long-Context Sparse Attention
arXiv 2608.19920 · 2026-09-29机器人学习中隐式动作的关键影响因素
What Matters for Latent Actions in Robot Learning
arXiv 2608.19613 · 2026-09-29对终身学习的两种思考：探索神经模型中的半球冗余与特化
In Two Minds about Lifelong Learning: Exploring Hemispheric Redundancy and Specialisation in Neural Models
arXiv 2608.19514 · 2026-09-29通过连续偏序集构建时空：旧基础与新发展
Spacetimes via Continuous posets: Old foundations and new developments
arXiv 2608.19502 · 2026-09-29超越多模态对齐：通过响应替换与有序执行验证物理语言
Beyond Multimodal Alignment: Shared Physical Representations Across Sensors and Action Orders
arXiv 2608.19492 · 2026-09-292026年ChildSafeAds共享任务：面向儿童的YouTube视频中的商业内容
ChildSafeAds Shared Task 2026: Commercial Content in Child-Facing YouTube Videos
arXiv 2608.19165 · 2026-09-29AI后训练AI缺失了什么？一项实证分析
What is Missing from AI Post-Training AI: An Empirical Analysis
arXiv 2608.19072 · 2026-09-29通过GauXC在CP2K中实现机器学习的Skala交换-关联泛函的分子级应用
Molecular Implementation of the Machine-Learned Skala Exchange-Correlation Functional in CP2K through GauXC
arXiv 2608.19033 · 2026-09-29RoboEdit：将人类操作视频转化为可扩展的机器人体验
RoboEdit: Turning Human Manipulation Videos into Scalable Robot Experience
arXiv 2608.18948 · 2026-09-29重整化群变换的簇表示及远离临界点的伊辛模型RG流收敛到平凡不动点的严格证明
Cluster Representation of Renormalization Group Transformations and a Rigorous Proof for Convergence of the RG-Flow of the Ising Model to Trivial Fixed Points away from Criticality
arXiv 2608.18862 · 2026-09-29基于潜世界模型的强化规划
Reinforced Planning with Latent World Models
arXiv 2608.18669 · 2026-09-29VA-Judger：用于联合视频-音频生成的人类偏好反馈奖励建模
VA-Judger: Reward Modeling from Human Preference Feedback for Joint Video-Audio Generation
arXiv 2608.18607 · 2026-09-29WhiteMatter：通过KV混合实现全对全跨层连接
WhiteMatter: All-to-All Cross-Layer Connections via KV Source Mixing
arXiv 2608.18486 · 2026-09-29划分支撑集，重建残差：用于视频生成和世界模型的无训练稀疏注意力
Partition the Support, Reconstruct the Residual: Training-Free Sparse Attention for Video Generation and World Models
arXiv 2608.18484 · 2026-09-29存在时变中介变量的双重差分模型
Difference-in-Differences Models in the Presence of Time-Varying Mediators and Lagged outcomes
arXiv 2608.18375 · 2026-09-29将配对自由费米子高斯态嵌入Gutzwiller投影的巴丁-库珀-施里弗（BCS）波函数
Embedding Paired Free-Fermion Gaussian States into Gutzwiller-Projected Bardeen--Cooper--Schrieffer Wave Functions
arXiv 2608.18168 · 2026-09-29对称稳定自回归序列的近单位根持续性
Near-unit-root persistence of symmetric and asymmetric stable autoregressive sequences
arXiv 2608.17927 · 2026-09-29与交叉多面体的Banach-Mazur距离的5/8下界
A $5/8$ Lower Bound on the Banach--Mazur Distance to the Cross-Polytope
arXiv 2608.17743 · 2026-09-29PTXBench：用于结合架构特定PTX优化GPU内核的大型语言模型基准测试与适配
PTXBench: Benchmarking and Adapting LLMs for GPU Kernel Optimization with Architecture-specific PTX
arXiv 2608.17379 · 2026-09-29一个奇异的$S^2\times S^2$与一个奇异的$\mathbb{CP}^2\\#\overline{\mathbb{CP}}^2$
An exotic $S^2\times S^2$ and an exotic $\mathbb{CP}^2\#\overline{\mathbb{CP}}^2$
arXiv 2608.17267 · 2026-09-29双Jaynes-Cummings模型中量子导引的动力学保护与保真度动力学
Dynamical protection of quantum steering and fidelity dynamics in the double Jaynes-Cummings model
arXiv 2608.17217 · 2026-09-29
