2026-09论文中文摘要
CyFA：具有相对时间分区内存的线性序列建模
CyFA: Linear Sequence Modeling with Relative-Time-Partitioned Memory
arXiv 2609.36259 · 2026-09-30kagome磁体HoV$_6$Sn$_6$的磁性质与费米面拓扑
Magnetic properties and Fermi-surface topology of kagome magnet HoV$_6$Sn$_6$
arXiv 2609.36258 · 2026-09-30当对冲改变收益：具有价格冲击和执行成本的美式期权复制
When Hedging Changes the Payoff: Option Replication with Price Impact and Execution Costs
arXiv 2609.36257 · 2026-09-30$S^2$ 上的代数向量场是 $2$-生成的
Algebraic Vector Fields on $S^2$ are $2$-generated
arXiv 2609.36256 · 2026-09-30具有C4v对称性的石墨烯圆盘谐振器方阵中模式的杂化、简并和偏振无关性
Hybridization, degeneracy and polarization independence of modes in square array of graphene disk resonators with symmetry C4v
arXiv 2609.36255 · 2026-09-30缓解大型推理模型中的欺骗性安全对齐
Towards Mitigating Deceptive Safety Alignment in Large Reasoning Models
arXiv 2609.36254 · 2026-09-30群体保真度：评估大语言模型中的群体代表性
Population Fidelity: Evaluating Population Representativeness in LLMs
arXiv 2609.36253 · 2026-09-30凝胶限域滚环扩增实现灵敏的单细胞蛋白形式分析
Gel-Confined Rolling-Circle Amplification Enables Sensitive Single-Cell Proteoform Analysis
arXiv 2609.36251 · 2026-09-30动作分块近端策略优化与反馈校正
Action Chunking Proximal Policy Optimization with Feedback Correction
arXiv 2609.36250 · 2026-09-30LDPC Fractus 码：具有递归结构的稀疏码
LDPC Fractus Codes: Sparse Codes with Recursive Structure
arXiv 2609.36249 · 2026-09-30拓扑阻碍无节点激子
Topology Obstructs Nodeless Excitons
arXiv 2609.36248 · 2026-09-30DER-Fault：面向不同运行条件下分布式能源集成配电系统的基于仿真的故障数据集
DER-Fault: A Simulation-Based Fault Dataset for DER-Integrated Distribution Systems Under Diverse Operating Conditions
arXiv 2609.36247 · 2026-09-30CoRe：协同进化奖励模型以缓解视频扩散模型中的潜在奖励黑客问题
CoRe: Co-Evolving Reward Models for Mitigating Latent Reward Hacking in Video Diffusion Models
arXiv 2609.36245 · 2026-09-30次要结构差异调节光系统II能量传递网络中天线亚基的功能特化
Minor Structural Differences Tune Functional Specialization of Antenna Subunits in the PSII Energy Transfer Network
arXiv 2609.36244 · 2026-09-30先修复再思考：基于笔画引导的风险感知满文手稿修复
Think Before You Restore: Risk-Aware Manchu Manuscript Restoration with Stroke-Guided Attention
arXiv 2609.36243 · 2026-09-30维度彩色Helly定理及拓扑变体
Dimensional colorful Helly theorems and topological variants
arXiv 2609.36242 · 2026-09-30拮抗肌腱驱动的双向操作灵巧机器人手的设计与验证
Design and Validation of an Antagonistic Tendon-Driven Dexterous Robotic Hand with Bidirectional Operation
arXiv 2609.36241 · 2026-09-30表征学习能否与损失最小化解耦？极化更新给出了答案
Can Representation Learning Decouple from Loss Minimization? Polar Updates Have an Answer
arXiv 2609.36240 · 2026-09-30认知专家语言模型更好地与相应的大脑系统对齐
Cognitive Expert Language Models Better Align with the Corresponding Brain Systems
arXiv 2609.36239 · 2026-09-30ChronoSRL：用于自监督强化学习的时序几何
ChronoSRL: Temporal Geometry for Self-Supervised Reinforcement Learning
arXiv 2609.36238 · 2026-09-30横向池化神经网络
Transversal Pooling Neural Networks
arXiv 2609.36237 · 2026-09-30几乎线性时间内最大化社交影响力
Maximizing Social Influence in Almost Linear Time
arXiv 2609.36236 · 2026-09-30Kaluza-Klein引力子级联衰变率的普适公式
A Universal Kaluza-Klein Graviton Cascade Rate
arXiv 2609.36234 · 2026-09-30两个二次曲面交集的Fano直线簇的Motivic类
Motivic classes of Fano schemes of lines on intersections of two quadrics
arXiv 2609.36233 · 2026-09-30Langevin动力学中自谐波漂移导致随机局域化密度底层域紧致化
Autonomous self-harmonic drift in Langevin dynamics causes a compaction of underlying domain for the density of Stochastic Localization
arXiv 2609.36232 · 2026-09-30补丁觅食环境中的社会认知模型：模型选择与参数可辨识性方法的案例研究
Socio-cognitive models in a patch foraging setting: a case study for model selection and parameter identifiability methods
arXiv 2609.36231 · 2026-09-30前馈层在Transformer动力学中的作用
The Role of Feed-Forward Layers in Transformer Dynamics
arXiv 2609.36230 · 2026-09-30评估H.E.S.S.和CTAO对脉冲星晕的探测能力与识别
Assessing the detectability and identification of pulsar halos with H.E.S.S. and CTAO
arXiv 2609.36229 · 2026-09-30欧盟人工智能法案合规检查器的实证研究与评估
An Empirical Study and Assessment of EU AI Act Compliance Checkers
arXiv 2609.36228 · 2026-09-30一步式下一潜在预测并非世界模型
One-Step Next-Latent Prediction Is Not a World Model
arXiv 2609.36227 · 2026-09-30Tachibana定理再探
Tachibana's Theorem Revisited
arXiv 2609.36226 · 2026-09-30提供使用约束可解性查询的3D障碍赛道游戏快速设计反馈
Providing Rapid Design Feedback for 3D Obstacle Course Games Using Constrained Solvability Queries
arXiv 2609.36225 · 2026-09-30相互对抗自训练与演化数据用于统一多模态模型
Mutually Adversarial Self-Training with Evolving Data for Unified Multimodal Models
arXiv 2609.36224 · 2026-09-30动态热定额在加速负荷增长下对配电变压器的益处
Benefits of Dynamic Thermal Rating for Distribution Transformers Under Accelerated Load Growth
arXiv 2609.36223 · 2026-09-30BASE：基于预测移除误差的批量感知专家选择，用于高效MoE解码
BASE: Batch-Aware Selection of Experts Using Predicted Removal Error for Efficient MoE Decoding
arXiv 2609.36222 · 2026-09-30语言模型在串行需求下如何差异性地重新分配注意力头活动
How Language Models Differ in Redistributing Attention-Head Activity Under Serial Demand
arXiv 2609.36221 · 2026-09-30典型范数的Hadwiger--Nelson问题
Hadwiger--Nelson Problem for Typical Norms
arXiv 2609.36220 · 2026-09-30LeRF：学习参考坐标系以进行视角推理
LeRF: Learning Reference Coordinate Frames for Perspective Taking Reasoning
arXiv 2609.36219 · 2026-09-30CineSubBench：从多语言电影字幕评估大语言模型的长篇叙事与文化理解能力
CineSubBench: Evaluating LLMs on Long-Form Narrative and Cultural Understanding from Multilingual Movie Subtitles
arXiv 2609.36218 · 2026-09-30预条件物理信息神经算子训练
Preconditioned Physics-Informed Neural Operator Training
arXiv 2609.36216 · 2026-09-30EnergyEminence：物理接地电网数字孪生中的源感知环境校准与评估
EnergyEminence: Source-Aware Environmental Calibration and Evaluation in a Physics-Grounded Grid Digital Twin
arXiv 2609.36215 · 2026-09-30迷失在翻译中：衡量非母语英语对大型语言模型终端用户表现的影响
Lost in Translation: Measuring the Effect of Non-Native English on End User Performance of Large Language Models
arXiv 2609.36214 · 2026-09-30群代数上的双变量自行车码
Bivariate Bicycle Codes over Group Algebras
arXiv 2609.36213 · 2026-09-30三维非完整车辆在球坐标下的逆最优反馈镇定
Inverse Optimal Feedback Stabilization of 3D Nonholonomic Vehicles in Spherical Coordinates
arXiv 2609.36212 · 2026-09-30零膨胀并非免费：Natário曲速驱动中欧拉负能量的尖锐最小值
Zero expansion is not free: the sharp minimum of Eulerian negative energy in Natário warp drives
arXiv 2609.36211 · 2026-09-30生成式去噪器雅可比矩阵的谱性质研究
On the spectral properties of generative denoiser Jacobians
arXiv 2609.36210 · 2026-09-30规范顺序问题：当大型语言模型作为多值关系的不可靠知识库时
The Canonical Order Problem: When Large Language Models Are Unreliable Knowledge Bases for Multi-Valued Relations
arXiv 2609.36209 · 2026-09-30可表示但未学习：编码秩与交互预测下限
Representable but Unlearned: Encoding Rank and the Interaction-Prediction Floor
arXiv 2609.36208 · 2026-09-30物理学是为了什么？为什么经典物理学不是量子力学的极限
What is Physics For? Why Classical Physics is not the Limit of Quantum Mechanics
arXiv 2609.36207 · 2026-09-30酷热海王星的凉爽多云之夜：LTT 9779 b 的泛波段 NIRISS/NIRSpec 透射光谱
Cool, Cloudy Nights on an Ultrahot Neptune: A Panchromatic NIRISS/NIRSpec Transmission Spectrum of LTT 9779 b
arXiv 2609.36206 · 2026-09-30
