2026-09论文中文摘要
ForVis：用于森林冠层下无人机飞行的VIO实地数据集与基准
ForVis: An In-Field Dataset and Benchmark for VIO Using Under-Canopy UAV Flights in Forests
arXiv 2609.35482 · 2026-09-29TopoEP：面向专家并行MoE训练的拓扑感知负载均衡
TopoEP: Topology-Aware Load Balancing for Expert-Parallel MoE Training
arXiv 2609.35481 · 2026-09-29从开放量子点共隧穿中提取任意子交换相位
Extracting the anyonic exchange phase from co-tunneling through an open quantum dot
arXiv 2609.35480 · 2026-09-29机器人工具设计从零开始：基于行为感知的层次优化
Robot Tool Design from Scratch via Behavior-Aware Hierarchical Optimization
arXiv 2609.35479 · 2026-09-29模拟计算再探：用于超声检测的全模拟极简损伤探测器，实现材料集成结构健康监测
Analog Computing revisited: A fully analog and minimalistic Damage Detector for Ultrasonic Testing enabling Material-Integrated Structural Health Monitoring
arXiv 2609.35478 · 2026-09-29原子级薄晶体银纳米结构中的等离激元增强二次谐波产生
Plasmon-Enhanced Second-Harmonic Generation in Atomically Thin Crystalline Silver Nanostructures
arXiv 2609.35477 · 2026-09-29自发上下文恢复：语言模型如何从受损输入中恢复
Spontaneous Context Restoration: How Language Models Recover from Corrupted Inputs
arXiv 2609.35475 · 2026-09-29一类 $\mathbb{C}$-凸域的强线性凸穷举
Strongly linearly convex exhaustion of a class of $\mathbb{C}$-convex domains
arXiv 2609.35474 · 2026-09-29手写文本识别存在于高像素方差子空间中
Handwritten Text Recognition Lives in the High-Pixel Variance Subspace
arXiv 2609.35473 · 2026-09-29为什么确定性PRM引导在离散扩散推理中表现不佳
Why Deterministic PRM Guidance Underperforms in Discrete Diffusion Reasoning
arXiv 2609.35472 · 2026-09-29深度学习探测到的 $z \sim 3.4$ 原星系团中的多样金属丰度
Diverse Metallicities in Protoclusters at $z \sim 3.4$ detected by Deep Learning
arXiv 2609.35471 · 2026-09-29重新思考流匹配中的因果动作标记化与条件退火
Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching
arXiv 2609.35469 · 2026-09-29探索NJL模型中磁场下由介子屏蔽质量揭示的$U_A(1)$对称性恢复
Exploring $U_A(1)$ symmetry restoration in a magnetic field by meson screening masses in the NJL model
arXiv 2609.35468 · 2026-09-29正特征中的拟$F$-分裂本原辛簇
Quasi-$F$-split primitive symplectic varieties in positive characteristic
arXiv 2609.35467 · 2026-09-29Tetra：以每参数2.7比特服务李奇格点量化的大语言模型
Tetra: Serving Leech-Lattice Quantized LLMs at 2.7 Bits per Parameter
arXiv 2609.35465 · 2026-09-29W2Rep：通过观察世界变化学习视觉表征
W2Rep: Learning Visual Representations by Watching the World Change
arXiv 2609.35464 · 2026-09-29A.D.A.M.O.（具多模态观测的语言驱动行动智能体）：面向虚拟人类的视觉-符号框架
A.D.A.M.O. (Agent for language-Driven Actions with Multimodal Observations): A Visual-Symbolic Framework for Virtual Humans
arXiv 2609.35463 · 2026-09-29CLIMB：通过多轮对话诊断共病情况的临床多病种基准
CLIMB: A Clinical Multimorbidity Benchmark for Diagnosing Co-occurring Conditions through Multiturn Conversations
arXiv 2609.35462 · 2026-09-29AraDynFact：阿拉伯语事实知识的动态评估
AraDynFact: Dynamic Evaluation of Factual Knowledge in Arabic
arXiv 2609.35461 · 2026-09-29压缩大语言模型重编程用于车载网络中的视觉辅助波束预测
Compressed LLM Reprogramming for Vision-Aided Beam Prediction in Vehicular Networks
arXiv 2609.35459 · 2026-09-29小-$x$ 下的微分希格斯产生
Differential Higgs production at small-$x$
arXiv 2609.35458 · 2026-09-29我们距离移除视觉编码器还有多远？无编码器多模态预训练的缩放定律
How Far Are We from Removing the Visual Encoder? Scaling Laws for Encoder-Free Multimodal Pretraining
arXiv 2609.35457 · 2026-09-29AutoBCI：基于脑电的脑机接口的预测引导智能体神经架构发现
AutoBCI: Forecast-Guided Agentic Neural Architecture Discovery for EEG-Based Brain--Computer Interfaces
arXiv 2609.35456 · 2026-09-29Regge 在奇境中
Regge in Wonderland
arXiv 2609.35455 · 2026-09-29流形稳定的流匹配
Manifold-Stable Flow Matching
arXiv 2609.35454 · 2026-09-29Courant-Hilbert 形变的引力实现
A gravitational realization of Courant-Hilbert deformations
arXiv 2609.35452 · 2026-09-29群论Zhang-Yeung不等式的一个简单证明
A simple proof of the group-theoretic Zhang-Yeung inequality
arXiv 2609.35451 · 2026-09-29从内部表征到通过预测误差改进模型
From internal representations to model improvement through prediction errors
arXiv 2609.35449 · 2026-09-29AI 匹配在双盲随机试验中改善难民就业
AI-based matching improves refugee employment in a double-blind randomized trial
arXiv 2609.35448 · 2026-09-29玻色子采样线性统计量估计的高效经典算法
Efficient classical algorithm for estimating linear statistics of Boson Sampling
arXiv 2609.35447 · 2026-09-29全静态架构中通过集体耗散实现固定时间高斯态转移
Fixed-Time Gaussian State Transfer via Collective Dissipation in a Fully Static Architecture
arXiv 2609.35446 · 2026-09-29DiMoP：基于帧级伪分类的扩散驱动运动表示学习用于基于骨架的动作识别
DiMoP: Diffusion-Driven Motion Representation Learning With Frame-Level Pseudo-Classification for Skeleton-Based Action Recognition
arXiv 2609.35444 · 2026-09-29极大性与实代数曲线上的构型空间
Maximality and configuration spaces on real algebraic curves
arXiv 2609.35442 · 2026-09-29Riccati 状态空间模型：非线性序列建模的非迭代并行化
Riccati State Space Models: Non-iterative Parallelization for Nonlinear Sequence Modeling
arXiv 2609.35441 · 2026-09-29SOLO：利用共享输出局部学习预训练十亿参数语言模型
SOLO: Pretraining Billion-Parameter Language Models with Shared-Output Local Learning
arXiv 2609.35440 · 2026-09-29修订而非重启：用于闭环世界-动作模型的可修订视觉规划
Revision, Not Restart: Revisable Visual Plans for Closed-Loop World-Action Models
arXiv 2609.35439 · 2026-09-29关于对数容量与牛顿容量的 Pólya 和 Szegő 猜想
Toward Pólya and Szegő's conjecture for logarithmic vs Newtonian capacity
arXiv 2609.35438 · 2026-09-29乳腺癌患者腋窝淋巴结转移的集成几何量化与形态分析框架
An integrated geometric quantification and shape analysis framework for axillary lymph node metastasis in breast cancer patients
arXiv 2609.35437 · 2026-09-29通过双束的轨道映射空间
Orbifold Mapping Spaces via Bibundles
arXiv 2609.35435 · 2026-09-29ReSPO：用于离线策略学习中梯度饥饿问题的重塑序列策略优化
ReSPO: Reshaped Sequence Policy Optimization for Gradient Starvation in Off-Policy Learning
arXiv 2609.35433 · 2026-09-29自进化编码智能体：从数字程序到物理世界智能
Self-Evolving Coding Agents: From Digital Programs to Physical-World Intelligence
arXiv 2609.35432 · 2026-09-29天空中的记忆：基于多智能体记忆聚合的低空问答
Memory in the Sky: Low-Altitude Question Answering with Multi-Agent Memory Aggregation
arXiv 2609.35431 · 2026-09-29条件均值算子的多任务学习：在动力系统与不确定性量化中的应用
Multi-Task Learning of Conditional Mean Operators: applications to dynamical systems and uncertainty quantification
arXiv 2609.35429 · 2026-09-29阵发性对流涌现的降阶耦合振子模型
A Reduced-Order Coupled Oscillator Model for the Emergence of Episodic Convection
arXiv 2609.35428 · 2026-09-29LLMs是通用异步智能体
LLMs are General Asynchronous Agents
arXiv 2609.35427 · 2026-09-29前沿学习：在能力边缘训练LLM推理器
Frontier Learning: Training LLM Reasoners at the Edge of Capability
arXiv 2609.35426 · 2026-09-29用于LLM解码的语义前缀预言机：契约与差分验证
Semantic Prefix Oracles for LLM Decoding: Contracts and Differential Validation
arXiv 2609.35425 · 2026-09-29基于频域二次约束的电力系统可扩展小信号稳定性评估
Scalable Small-Signal Stability Assessment of Power Systems Based on Frequency-Domain Quadratic Constraints
arXiv 2609.35424 · 2026-09-29JWST/NIRSpec 密挡板光谱中的源-谱线关联
Source--Line Associations in JWST/NIRSpec Dense-Shutter Spectroscopy
arXiv 2609.35423 · 2026-09-29V405 Draconis：一个具有第三伴星和四合星系统候选体的深度相接双星
V405 Draconis: A Deep Contact Binary with a Tertiary Companion and a Quadruple System Candidate
arXiv 2609.35422 · 2026-09-29
