2026-09论文中文摘要
mHC如何利用其残差流？选择性路由与近恒等混合
How Does mHC Use Its Residual Streams? Selective Routing and Near-Identity Mixing
arXiv 2609.05309 · 2026-09-07用于可扩展贝叶斯矩条件推断的子采样伪后验
Subsampled Pseudo-posteriors for Scalable Bayesian Moment-condition Inference
arXiv 2609.05308 · 2026-09-07基于预报集合的区间数据同化主动二元阈值查询设计
Forecast-Ensemble-Based Active Binary-Threshold Query Design for Interval Data Assimilation
arXiv 2609.05307 · 2026-09-07高能QCD演化中的混沌是与目标无关的
Chaos is Target-Blind in High-Energy QCD Evolution
arXiv 2609.05306 · 2026-09-07冗余表征为何能记忆？谱系隔离而非掩码
What Makes a Redundant Representation Remember? Lineage Isolation, Not Masking
arXiv 2609.05304 · 2026-09-07高光谱图像超分辨率的空间-光谱细化学习与互补观测校准
Learning Spatial-Spectral Refinement and Calibrating Complementary Observations for Hyperspectral Image Super-Resolution
arXiv 2609.05303 · 2026-09-07关于Cui、Gu和Tang提出的18着色广义Frobenius分拆猜想的证明
Proof of a Conjecture of Cui, Gu and Tang on 18-Colored Generalized Frobenius Partitions
arXiv 2609.05302 · 2026-09-07詹姆斯·韦布空间望远镜（JWST）排除了岩质温带系外行星周围半径小至0.1倍地球半径的系外卫星
JWST Excludes Exomoons Down to 0.1 Earth Radii Around a Rocky, Temperate Exoplanet
arXiv 2609.05301 · 2026-09-07用于密集不确定人群中机器人导航的人-人及人-机器人交互Transformer（H2INT）
Human-Human & Human-Robot Interaction Transformer (H2INT) for Robot Navigation in Dense and Uncertain Crowds
arXiv 2609.05300 · 2026-09-07扰动指数的一个改进上界
An Improved Upper Bound for the Perturbation Index
arXiv 2609.05299 · 2026-09-07合作多智能体强化学习的在线变点检测
Online Change-point Detection for Cooperative Multi-Agent Reinforcement Learning
arXiv 2609.05298 · 2026-09-07随机正则有向图的定向Kesten--McKay定律
The oriented Kesten--McKay law for random regular digraphs
arXiv 2609.05297 · 2026-09-07LexFlip：用于法律含义保留指标的解离诊断工具
LexFlip: A Dissociation Diagnostic for Legal Meaning Preservation Metrics
arXiv 2609.05296 · 2026-09-07RISE：基于自外推策略蒸馏的递归改进
RISE: Recursive Improvement via Self-Extrapolating Policy Distillation
arXiv 2609.05295 · 2026-09-07从变分自编码器（VAE）的误差中学习以支持基于心电图（ECG）的心肌瘢痕鉴别诊断
Learning from VAE Errors to support ECG-based Differential Diagnosis of Myocardial Scar
arXiv 2609.05294 · 2026-09-07通过加权能量耗散得到的四阶退化抛物型方程的解
Solutions to fourth order degenerate parabolic equations obtained via weighted energy dissipation
arXiv 2609.05293 · 2026-09-07具有双传递置换自同构群的MDS码的长度
On the lengths of MDS codes with a two-transitive permutation automorphism group
arXiv 2609.05292 · 2026-09-07从另一侧看非弹性过程：基于LZ高反冲事件的氙激发信号
Inelastic from the Other Side: Xenon Excitation Signals in Light of the LZ High-Recoil Event
arXiv 2609.05291 · 2026-09-07范围更新网络中的通用二次度定律
Universal Exponent-Two Degree Laws in Range-Renewal Networks
arXiv 2609.05290 · 2026-09-07超越聚合分数：评估基于参考的自动评估方法的行为正确性假设
Beyond Aggregate Scores: Behavioral Correctness Assumptions for Assessing Reference-Based Automatic Evaluation Methods
arXiv 2609.05289 · 2026-09-07Augur：预测关系数据存储应用中的视图可串行性违反
Augur: Predicting View Serializability Violations in Relational Data Store Applications
arXiv 2609.05288 · 2026-09-07与库耦合的光子玻色-爱因斯坦凝聚体的涨落：用自由能模型描述相干性质
Fluctuations of a Photon Bose-Einstein Condensate Coupled to a Reservoir: Describing Coherence Properties in a Free-Energy Model
arXiv 2609.05287 · 2026-09-07目标未知前的统计约简：两个边界结果
Statistical reduction before the target is known: two boundary results
arXiv 2609.05286 · 2026-09-07红移2.2处原星系团核心的快速淬灭与早期气体耗尽
Rapid quenching and early gas depletion in the core of a galaxy protocluster at z=2.2
arXiv 2609.05285 · 2026-09-07GUT：基于图复杂度量化与优化大语言模型的推理不确定性
GUT: Quantifying and Optimizing the Reasoning Uncertainty of LLMs via Graph Complexity
arXiv 2609.05284 · 2026-09-07边界条件小Robin扰动下电压势的渐近展开
Asymptotic expansion of the voltage potential under small Robin perturbations of its boundary conditions
arXiv 2609.05283 · 2026-09-07用于意图感知的机器人-人类双手交接的时间触觉编码与柔顺性
Temporal Tactile Encoding and Compliance for Intent-Aware Robot-to-Human Bimanual Handover
arXiv 2609.05282 · 2026-09-07KanAdapter：一种基于柯尔莫哥洛夫-阿诺尔德网络的即插即用模块，用于高效微调基础语音模型
KanAdapter: A Kolmogorov-Arnold Network-based Plug-and-Play Module for Efficient Fine-tuning of Foundation Speech Models
arXiv 2609.05281 · 2026-09-07具有射影对称性的二维度量的Sophus Lie问题：局部分类的完成
Sophus Lie's problem on two-dimensional metrics with projective symmetries: completing the local classification
arXiv 2609.05280 · 2026-09-07测试大语言模型（LLM）智能体团队的可互换性
Testing Interchangeability in LLM Agent Teams
arXiv 2609.05279 · 2026-09-07Ritgard：GitHub上社会技术制品的主题岛屿
Ritgard: T(r)opical Islands of Socio-Technical Artifacts on GitHub
arXiv 2609.05278 · 2026-09-07单位球上的k-Hessian特征值的渐近行为
Asymptotics for the $k$-Hessian Eigenvalue on the Unit Ball
arXiv 2609.05277 · 2026-09-07将异质动力学与局部熵关联起来
Connecting heterogeneous dynamics with local entropy
arXiv 2609.05276 · 2026-09-07不要丢弃Dropout：优化层稀疏性以实现高效的大语言模型训练与推理
Don't Drop Dropout: Optimizing Layer Sparsity for Efficient LLM Training and Inference
arXiv 2609.05275 · 2026-09-07如何推测智能体编码中的不确定性？一种草稿模型门控方法
How to Speculate about Uncertainty in Agentic Coding? A Draft-Model Gate Method
arXiv 2609.05274 · 2026-09-07金属-半导体渗流复合材料的热电性质
Thermoelectric properties of metal-semiconductor percolation composites
arXiv 2609.05273 · 2026-09-07ROCKETS I：研究火箭效应对两个被红外泡包裹的邻近年轻疏散星团的影响
ROCKETS I: Investigating the Impact of the Rocket Effect on Two Nearby Young Open Clusters Encased in Infrared Bubbles
arXiv 2609.05272 · 2026-09-07相对（τ）、扩张器与某些扩张映射的相关性衰减
Relative ($τ$), Expanders, and Decay of Correlations for certain Expanding Maps
arXiv 2609.05271 · 2026-09-07面向计算设计科学的人工智能：负责任的人机协同框架及短视频安全监控案例研究
AI for Computational Design Science: A Responsible Human-AI Framework and Case Study on Short-Form Video Safety Surveillance
arXiv 2609.05270 · 2026-09-07CONTINUITY：用于可组合LLM智能体控制的安全-上下文契约
CONTINUITY: Security-Context Contracts for Composable LLM Agent Controls
arXiv 2609.05269 · 2026-09-07极大（外）平面图的拉姆齐性质
Ramsey properties of maximal (outer)planar graphs
arXiv 2609.05268 · 2026-09-07终端激波处的能量分配
Energy Partitioning at the Termination Shock
arXiv 2609.05267 · 2026-09-07TacPAC：面向接触丰富型操作的世界-动作模型中的触觉预测与实时动作校正
TacPAC: Tactile Prediction and Real-Time Action Correction in World-Action Models for Contact-Rich Manipulation
arXiv 2609.05266 · 2026-09-0710颗系外行星的JWST MIRI透射光谱的同质巡天
A Homogeneous Survey of JWST MIRI Transmission Spectra of 10 Exoplanets
arXiv 2609.05264 · 2026-09-07用于快速大规模系统发育推理的自监督词汇表示学习
Self-Supervised Lexical Representation Learning for Fast, Large-Scale Phylogenetic Inference
arXiv 2609.05262 · 2026-09-07Trace2Tower：面向大语言模型智能体的多级技能的转换感知特征轨迹诱导方法
Trace2Tower: Transition-Aware EigenTrace Induction of Multi-Level Skills for LLM Agents
arXiv 2609.05261 · 2026-09-0714是使得所有立方图的每个4-全着色均为均匀着色的最大阶数
Order 14 is the largest order for which every 4-total coloring of every cubic graph is equitable
arXiv 2609.05259 · 2026-09-07计算机视觉中的常识推理：基础、最新进展与未来方向
Commonsense Reasoning in Computer Vision: Foundations, Recent Advancements, and Future Directions
arXiv 2609.05257 · 2026-09-07受控情绪刺激下基于机器学习算法的生理信号情绪识别
Emotion Recognition from Physiological Signals Using Machine Learning Algorithms Under Controlled Emotional Stimuli
arXiv 2609.05256 · 2026-09-07用于高效高斯溅射的紧凑神经外观模型
Compact Neural Appearance Models for Efficient Gaussian Splatting
arXiv 2609.05255 · 2026-09-07
