2026-09论文中文摘要
OneLA：在生成式推荐中将线性注意力解码扩展到大规模束搜索
OneLA: Scaling Linear-Attention Decoding to Large Beams in Generative Recommendation
arXiv 2609.12399 · 2026-09-14The Core 中的小生境智能体
Niching Agents in The Core
arXiv 2609.12398 · 2026-09-14基于偏差原理的随机方差缩减梯度法在线性反问题中的早期停止
Early stopping of stochastic variance reduced gradient for linear inverse problems by the discrepancy principle
arXiv 2609.12396 · 2026-09-14高斯泼溅是否再次神经化？关于学习参数化的分类法与受控研究
Is Gaussian Splatting Becoming Neural Again? A Taxonomy and Controlled Study of Learned Parameterization
arXiv 2609.12395 · 2026-09-14量子卷积神经网络用于预测核电荷半径
Quantum convolutional neural network for predicting nuclear charge radii
arXiv 2609.12393 · 2026-09-14近场毫米波混合波束赋形系统中反馈高效的波束-用户关联
Feedback-Efficient Beam-User Association for Near-Field mmWave Hybrid Beamforming Systems
arXiv 2609.12392 · 2026-09-14$t$-相交置换族的极大 $t$-多样性
The maximum $t$-diversity of $t$-intersecting families of permutations
arXiv 2609.12391 · 2026-09-14赛弗特1型星系长期中红外颜色变化的系统性研究
A systematic study of the long-term mid-infrared color variations of Seyfert 1 galaxies
arXiv 2609.12390 · 2026-09-14通过并行概率比特动力学中的位对齐时间强化降低LDPC解码错误率
Error-Rate Reduction in LDPC Decoding via Bit-Aligned Temporal Reinforcement in Parallel Probabilistic-Bit Dynamics
arXiv 2609.12389 · 2026-09-14RF-VoID：通过窄带射频表征学习实现带宽高效的外墙瓷砖空鼓检测
RF-VoID: Towards Bandwidth-Efficient Exterior Tile Void Detection via Narrowband Radio-Frequency Representation Learning
arXiv 2609.12388 · 2026-09-14制裁会适得其反吗？支持乌克兰的宏观经济影响的新证据
Do Sanctions Backfire? New Evidence on the Macroeconomic Effects of Supporting Ukraine
arXiv 2609.12387 · 2026-09-14分割共形预测与标签偏移调整的贝叶斯分数
Split Conformal Prediction with Label-Shift-Adjusted Bayesian Scores
arXiv 2609.12386 · 2026-09-14Kac-Moody 群的 Hecke 代数的余中心
Cocenter of Hecke algebras of Kac-Moody groups
arXiv 2609.12385 · 2026-09-14理解多样任务目标与约束下的全身机器人遥操作策略
Understanding Whole-Body Robot Teleoperation Strategies Under Diverse Task Objectives and Constraints
arXiv 2609.12384 · 2026-09-14三维各向异性Navier-Stokes方程的能量等式
On energy equality of the 3D anisotropic Navier-Stokes equations
arXiv 2609.12383 · 2026-09-14基于表示的掩码扩散模型
Representation-based Masked Diffusion Model
arXiv 2609.12382 · 2026-09-14宇宙射线空气簇射中首次相互作用二次谐波各向异性的演化
Evolution of first-interaction second-harmonic anisotropy in cosmic-ray air showers
arXiv 2609.12380 · 2026-09-14ForgeMegakernel：一种用于高效自回归模型解码Megakernel的通用框架
ForgeMegakernel: A General Framework for Efficient Auto-Regressive Model Decode Megakernels
arXiv 2609.12379 · 2026-09-14适用于M87*类吸积流的理想GRMHD模拟的局域化电流片磁扩散与加热方案
A Localized Current-Sheet Magnetic-Diffusion and Heating Prescription for Ideal-GRMHD Simulations of M87*-like Accretion Flows
arXiv 2609.12377 · 2026-09-14风险估计偏差通过差异检测的动态放大：基于历史协变量的马尔可夫模型
Dynamic Amplification of Risk-Estimate Bias Through Differential Detection: A Markov Model for History-Based Covariates
arXiv 2609.12376 · 2026-09-14ChronicleRec：面向终身用户建模的预训练时间锚定令牌
ChronicleRec: Pre-training Temporally Anchored Tokens for Lifelong User Modeling
arXiv 2609.12375 · 2026-09-14上下文感知的因果注视预测：用于车内跟踪丢失期间的人车交互
Context-Aware Causal Gaze Forecasting for Human-Vehicle Interaction During In-Cabin Tracking Dropouts
arXiv 2609.12374 · 2026-09-14面向LLM的鲁棒个性化对齐：缓解多轮对话中的角色漂移
Toward Robust Personalized Alignment for LLMs: Mitigating Persona Drift in Multi-Turn Dialogue
arXiv 2609.12373 · 2026-09-14正则多部竞赛图的4-弧泛圈性
4-Arc-Pancyclicity of Regular Multipartite Tournaments
arXiv 2609.12372 · 2026-09-14READ：学习风险知情场用于端到端自动驾驶
READ: Learning Risk-Informed Fields for End-to-End Autonomous Driving
arXiv 2609.12371 · 2026-09-14一种计算具有Robin边界条件的Schrödinger算子众多特征值及其渐近性的谱元方法
A spectral-element method for computing many eigenvalues and their asymptotics of the Schrödinger operator with Robin boundary condition
arXiv 2609.12369 · 2026-09-14$\mathbb{T}^m$-不变Toeplitz代数的局部化框架：Gelfand理论的应用
A localization framework for $\mathbb{T}^m$-Invariant Toeplitz Algebras: Applications to Gelfand Theory
arXiv 2609.12368 · 2026-09-14基于成对似然比的成员推断
Membership Inference via Pairwise Likelihood Ratios
arXiv 2609.12367 · 2026-09-14ORQA：面向LLM专业知识的职业现实问答框架
ORQA: An Occupation-Realistic Question and Answer Framework for LLM Professional Knowledge
arXiv 2609.12366 · 2026-09-14ICU警报的认证式AI分诊
Certified AI Triage of ICU Alarms
arXiv 2609.12365 · 2026-09-14LatentVerse：理解多模态潜在表示中共享与模态特定信息的框架
LatentVerse: A Framework for Understanding Shared and Modality-Specific Information in Multimodal Latent Representations
arXiv 2609.12364 · 2026-09-14利用非线性光学观测金刚石中的拓扑表面声子
Observation of topological surface phonons in diamond with nonlinear optics
arXiv 2609.12362 · 2026-09-14粗糙奇异积分算子的加权端点变差与跳跃不等式
Weighted Endpoint Variational and Jump Inequalites for Rough Singular Integrals
arXiv 2609.12361 · 2026-09-14一种证据优先的多大语言模型框架，用于可审计的关键基础设施依赖建模
An Evidence-First Multi-LLM Framework for Auditable Critical-Infrastructure Dependency Modeling
arXiv 2609.12360 · 2026-09-14通过声子辅助外差放大揭示ZrTe$_5$中隐藏的反演对称性破缺
Revealing Hidden Inversion Symmetry Breaking in ZrTe$_5$ via Phonon-Assisted Heterodyne Amplification
arXiv 2609.12359 · 2026-09-14Mie共振增强的混合光子结构手性响应
Enhanced chiral response of hybrid photonic structures empowered by Mie resonances
arXiv 2609.12358 · 2026-09-14近似对偶框架的不存在性：加权复合算子视角
Nonexistence of Approximately Dual Frames via Weighted Composition Operators
arXiv 2609.12357 · 2026-09-14当连接并不意味着相似：绘制SNAP-KG在流式实体集成中的同质性边界
When Connected Does Not Mean Similar: Charting the Homophily Boundary of SNAP-KG for Streaming Entity Integration
arXiv 2609.12356 · 2026-09-14Wasserstein空间的Gromov紧致化性
Gromov-compactifiability of Wasserstein spaces
arXiv 2609.12355 · 2026-09-14CueMem：用于长期对话记忆的线索引导式上下文重建
CueMem: Cue-Guided Context Reconstruction for Long-Term Conversational Memory
arXiv 2609.12354 · 2026-09-14SynthSentry：检测语言模型训练数据中的合成数据污染
SynthSentry: Detecting Synthetic Data Contamination in Language Model Training Data
arXiv 2609.12353 · 2026-09-14铁取代4d过渡金属硫化物中的涌现磁性、重电子与压力诱导的重入超导电性
Emergent magnetism, heavy electrons and pressure-induced reentrant superconductivity in the iron substituted 4d transition-metal sulfides
arXiv 2609.12351 · 2026-09-14EgoMaize：严重田间遮挡下的第一人称玉米实例分割基准
EgoMaize: A First-Person Maize Instance Segmentation Benchmark under Severe Field Occlusion
arXiv 2609.12350 · 2026-09-14一种用于机器人介导任务的可部署架构（DART）：在社交辅助机器人引导的认知行为疗法练习中的评估
A Deployable Architecture for Robot-Mediated Tasks (DART): Evaluation in Socially Assistive Robot-Guided Cognitive Behavioral Therapy Exercises
arXiv 2609.12349 · 2026-09-14e-Traceroute：用于无碳能源利用的物理可追踪电力路由
e-Traceroute: Physically traceable electricity routing for carbon-free energy utilization
arXiv 2609.12348 · 2026-09-14DWMP：利用双世界模型实现人形机器人跨越障碍
DWMP: Leveraging Dual World Models for Humanoid Obstacle Traversal
arXiv 2609.12347 · 2026-09-14低个体度测试在没有对角线测试时不具有量子可靠性
The Low-Individual-Degree Test Without the Diagonal-Lines Test Is Not Quantum-Sound
arXiv 2609.12346 · 2026-09-14ParaRecover：并行工具使用智能体中错误定位与恢复的过程级基准
ParaRecover: A Process-Level Benchmark for Error Localization and Recovery in Parallel Tool-Use Agents
arXiv 2609.12345 · 2026-09-14场协同与分形几何驱动的液冷板对流换热优化
Convective Heat Transfer Optimization for Liquid Cooling Plates Driven by Field Synergy and Fractal Geometry
arXiv 2609.12344 · 2026-09-14VS-Splat：用于稀疏视图端到端三维物体重建的体素选择性前馈高斯泼溅
VS-Splat: Voxel-Selective feed-forward Gaussian Splatting for end-to-end 3D object reconstruction from sparse-views
arXiv 2609.12343 · 2026-09-14
