2026-09论文中文摘要
中纬度低层电离层太阳耀斑期间电子密度与响应时间延迟的数值分析
Numerical analysis of electron density and response time delay during solar flares in mid-latitudinal lower ionosphere
arXiv 2609.34361 · 2026-09-29CoeF-SFL：保留协作式客户端-服务器学习并增强通信效率
CoeF-SFL: Preserving Collaborative Server-Client Learning with Enhanced Communication Efficiency
arXiv 2609.34360 · 2026-09-29通过运行时程序状态推理改进代码大语言模型
Improving Large Language Models for Code through Runtime Program-State Reasoning
arXiv 2609.34359 · 2026-09-29FORGE：冻结LLM智能体的有据证据的形式最优路由
FORGE: Form-Optimal Routing of Grounded Evidence for Frozen LLM Agents
arXiv 2609.34358 · 2026-09-29北冕座R型变星的变暗：尘埃气体云掩食模型
R Coronae Borealis fadings: a dusty gas cloud eclipse model
arXiv 2609.34357 · 2026-09-29预测性语义安全：从视觉物理推理到安全关键控制
Predictive Semantic Safety: From Visual Physical Reasoning to Safety-Critical Control
arXiv 2609.34356 · 2026-09-29利用高灵敏度环形微热电偶结观测巨大且可调谐的近场热辐射
Observing Colossal and Tunable Near-field Thermal Radiation with A Highly Sensitive Annular Micro-thermocouple Junction
arXiv 2609.34355 · 2026-09-29FAST-Brain：一种流对齐的时空代理脑模型
FAST-Brain: A Flow-Aligned Spatio-Temporal Surrogate Brain Model
arXiv 2609.34354 · 2026-09-29SemRD-V2X：面向协作感知的闭包引导通信与有界推理
SemRD-V2X: Closure-Guided Communication with Bounded Inference for Cooperative Perception
arXiv 2609.34353 · 2026-09-29第二大特征值并不约束稳态熵产生
Second largest eigenvalue does not bound stationary entropy production
arXiv 2609.34352 · 2026-09-29PolyCIM：基于多面体编译提升数字CIM加速器中的数据复用
PolyCIM: Improving Data Reuse in Digital CIM Accelerators with Polyhedral-Based Compilation
arXiv 2609.34351 · 2026-09-29不同脉冲持续时间真空二极管中产生的异常能量电子
Electrons with Anomalous Energy Generated in Vacuum Diodes with Different Pulse Duration
arXiv 2609.34350 · 2026-09-29用于保持因果性的合成表格数据生成的模糊分布建模
Fuzzy Distribution Modeling for Synthetic Tabular Data Generation with Causality Preservation
arXiv 2609.34349 · 2026-09-29交换子记忆：语言模型中的稀疏、路径局部读取与操控
Commutator Memory: Sparse, Path-Local Reading and Steering in Language Models
arXiv 2609.34348 · 2026-09-29SAIL：基于大语言模型的空间音频智能，通过解耦声学-空间编码与双流Q-Former实现
SAIL: Spatial Audio Intelligence with Large Language Models via Disentangled Acoustic-Spatial Encoding and Dual-Stream Q-Former
arXiv 2609.34347 · 2026-09-29E-WAVE：通过扭曲对齐视觉编码的事件驱动连续光流
E-WAVE: Event-based Continuous Optical Flow via Warping-Aligned Visual Encoding
arXiv 2609.34346 · 2026-09-29学习转向，转向以见：通过可训练向量揭示大型语言模型中RLVR的几何结构
Learning to Steer, Steering to See: Unveiling the Geometry of RLVR in Large Language Models via Trainable Vectors
arXiv 2609.34344 · 2026-09-29数据集蒸馏中的组合差距
The Composition Gap in Dataset Distillation
arXiv 2609.34343 · 2026-09-29SAGE：用于高效LLM游戏对弈的结构化策略推理
SAGE: Structured Strategic Reasoning for Efficient LLM Game Playing
arXiv 2609.34342 · 2026-09-29基于四边形单元的设计相关压力驱动柔顺机构的鲁棒拓扑优化比较研究
A Comparative Study on Robust Topology Optimization of Design-Dependent Pressure-Actuated Compliant Mechanisms with Quadrilateral Elements
arXiv 2609.34341 · 2026-09-29VarioPath：面向PCIe GPU集群的工作负载感知的全对全通信
VarioPath: Workload-Aware All-to-All Communication for PCIe GPU Clusters
arXiv 2609.34340 · 2026-09-29有向时间探索问题
The directed temporal exploration problem
arXiv 2609.34339 · 2026-09-29中子星修正和直接 Urca 冷却速率的中等介质效应
Medium effects on neutron star modified and direct Urca cooling rates
arXiv 2609.34338 · 2026-09-29音频令牌作为预算资源：用于可扩展音频表示的边际效用分配
Audio Tokens as a Budgeted Resource: Marginal-Utility Allocation for Scalable Audio Representations
arXiv 2609.34337 · 2026-09-29迈向星系模拟中恒星形成与星系风的分辨率收敛
Toward Resolution Convergence for Star Formation and Galactic Winds in Galaxy Simulations
arXiv 2609.34336 · 2026-09-29SkillPE：面向文本到视频提示工程的创意导向电影技能进化
SkillPE: Creativity-Oriented Cinematic Skill Evolution for Text-to-Video Prompt Engineering
arXiv 2609.34335 · 2026-09-29HOCCL：将集体通信从GPU核心卸载以加速分布式训练
HOCCL: Offloading Collective Communication from GPU Cores to Accelerate Distributed Training
arXiv 2609.34334 · 2026-09-29堆叠智能超表面-衍射深度神经网络用于星上SAR Level-0原始数据地形分类
Stacked Intelligent Metasurface-Diffractive Deep Neural Networks for Onboard Terrain Classification from SAR Level-0 Raw Data
arXiv 2609.34333 · 2026-09-29各向异性量子拉比模型中局域体态间的拓扑边缘态观测
Observation of a topological edge state among localized bulk states in the anisotropic quantum Rabi model
arXiv 2609.34332 · 2026-09-29最终高斯局部树上的Riesz变换
Riesz transform on eventually Gaussian local trees
arXiv 2609.34331 · 2026-09-29NGC6822演化研究：通过演化恒星族探究恒星形成历史
A Study on the Evolution of NGC6822. I. Investigating the Star Formation History through Evolved Stellar Populations
arXiv 2609.34329 · 2026-09-29从样本到形状：拓扑不变量的有限方法
From samples to shape: a finite approach to topological invariants
arXiv 2609.34328 · 2026-09-29知道何时思考不够：教小型推理模型超越其参数知识进行推理
Knowing When Thinking Is Not Enough: Teaching Small Reasoning Models to Reason Beyond Their Parametric Knowledge
arXiv 2609.34327 · 2026-09-29无需嵌入的路由：基于正则表达式的快速可解释路由
Routing Without Embeddings: Fast And Interpretable Routing With Regular Expressions
arXiv 2609.34326 · 2026-09-29DORA：用于视觉Transformer令牌剪枝的动态在线强化智能体
DORA: Dynamic Online Reinforcement Agent for Token Pruning in Vision Transformers
arXiv 2609.34325 · 2026-09-29遗迹中微子探测小尺度原初非高斯性
Relic Neutrinos Probing Small Scale Primordial Non-Gaussianity
arXiv 2609.34324 · 2026-09-29Gross-Neveu-Yukawa 模型中领先扭结算符的三圈反常维度
Three-loop anomalous dimensions of leading-twist operators in the Gross-Neveu-Yukawa model
arXiv 2609.34323 · 2026-09-29测试时扩展：基于预算约束的多属性验证
Test-Time Scaling via Budgeted Multi-Attribute Verification
arXiv 2609.34322 · 2026-09-29一次 rollout 足矣：GUI 智能体的完全测试时适应
One Rollout Is All You Get: Fully Test-Time Adaptation for GUI Agents
arXiv 2609.34321 · 2026-09-29LLM智能体评估的认证选择性自动化
Certified Selective Automation of LLM Agent Evaluation
arXiv 2609.34320 · 2026-09-29文本-视觉协同令牌缓存：一种用于高效视觉-语言-动作推理的无训练框架
Text-Vision Synergistic Token Caching: A Training-Free Framework for Efficient Vision-Language-Action Inference
arXiv 2609.34319 · 2026-09-29HyDra：在生产规模下揭秘并驯服动态上下文并行
HyDra: Demystifying and Taming Dynamic Context Parallelism at Production Scale
arXiv 2609.34318 · 2026-09-29开源项目中不完整软件变更经验数据集的构建
The Construction of an Empirical Dataset of Incomplete Software Changes from Open Source Projects
arXiv 2609.34317 · 2026-09-29动力学参数：时间序列基础模型的可解释性框架
Dynamical Parameters: An Interpretability Framework for Time-Series Foundation Models
arXiv 2609.34316 · 2026-09-29增强James--Stein估计用于高维主特征向量与特征子空间
Augmented James--Stein estimation for leading eigenvectors and eigenspaces in high dimensions
arXiv 2609.34315 · 2026-09-29PlaylistEval：视频语言裁判能否在日级及更长时间尺度上被信任？
PlaylistEval: Can Video-Language Judges Be Trusted at Day Scale and Beyond?
arXiv 2609.34314 · 2026-09-29ControlScope：LLM智能体中的工作流修订与可靠性
ControlScope: Workflow Revision and Reliability in LLM Agents
arXiv 2609.34313 · 2026-09-29Gauss--Manin 联络延拓到全纯余容许 D-cap-模
Gauss--Manin connections extend to holonomic coadmissible D-cap-modules
arXiv 2609.34312 · 2026-09-29粗糙极大奇异积分算子交换子的弱端点估计
Weak Endpoint Estimate for Commutators of Rough Maximal Singular Integral operators
arXiv 2609.34311 · 2026-09-29黎曼差分凸优化用于K-Means聚类
Riemannian Difference-of-Convex Optimization for K-Means Clustering
arXiv 2609.34310 · 2026-09-29
