2026-09论文中文摘要
费米子Fokker-Planck模型中定量时间渐近的熵水平方法——非饱和现象的出现
An entropy-level method for quantitative time-asymptotic in the Fokker-Planck model for fermions -- appearance of non saturation
arXiv 2609.33396 · 2026-09-29保留语素：面向尼泊尔语的形态学引导预分词
Preserving Morphemes: Morphology-Guided Pre-Tokenization for Nepali
arXiv 2609.33395 · 2026-09-29基于条件潜在去噪的跨模态翻译用于视频深度伪造检测
Cross-modal Translation via Conditional Latent Denoising for Video Deepfake Detection
arXiv 2609.33394 · 2026-09-29具有时空非局部衰减的双曲方程中多种时间依赖系数的同时恢复
Simultaneous recovery of various time-dependent coefficients in a hyperbolic equation with space-time nonlocal attenuation
arXiv 2609.33393 · 2026-09-29AutoHGNN：面向超图神经网络的鲁棒且高效的神经架构搜索
AutoHGNN: Robust and Efficient Neural Architecture Search for Hypergraph Neural Networks
arXiv 2609.33392 · 2026-09-29超越时间戳：面向长时程智能体的决策对齐在线策略蒸馏
Beyond Timestamps: Decision-Aligned On-Policy Distillation for Long-Horizon Agents
arXiv 2609.33391 · 2026-09-29从位置风险到块生存：扩散语言模型的更快生成
From Position Risks to Block Survival: Faster Generation for Diffusion Language Models
arXiv 2609.33390 · 2026-09-29强耦合量子平衡中的温度-哈密顿量歧义性
Temperature--Hamiltonian Ambiguity in Strong-Coupling Quantum Equilibrium
arXiv 2609.33389 · 2026-09-29单位范数紧框架稀疏坐标补全的秩稳定性
Rank Stabilization for Sparse Coordinate Completion of Unit-Norm Tight Frames
arXiv 2609.33388 · 2026-09-29从灰盒到绿盒：物理信息机器学习何时能减少结构健康监测中的碳足迹？
From Grey-Box to Green-Box: When can Physics-Informed Machine Learning Reduce Carbon Footprints in Structural Health Monitoring?
arXiv 2609.33387 · 2026-09-29SIVIA-RSI：图解技能的源基适应
SIVIA-RSI: Source-Grounded Adaptation of Diagramming Skills
arXiv 2609.33386 · 2026-09-29OLED-MoE：通过迭代间局部性感知专家卸载加速基于MoE的dLLM推理
OLED-MoE: Accelerating MoE-Based dLLM Inference via Inter-Iteration Locality-Aware Expert Offloading
arXiv 2609.33385 · 2026-09-29局部线性最小化预言机其实是一种投影方法！
Local LMO is Secretly a Projection Method!
arXiv 2609.33383 · 2026-09-29WideSWE：编码智能体能否跨仓库协调更改？
WideSWE: Can Coding Agents Coordinate Changes Across Repositories?
arXiv 2609.33382 · 2026-09-29群不变解在基态以下的散射：四阶非线性薛定谔方程
Scattering of group-invariant solutions below the ground state for the fourth-order NLS
arXiv 2609.33381 · 2026-09-29复杂系统的随机物理约束算子推断
Stochastic Physics-Constrained Operator Inference for Complex Systems
arXiv 2609.33380 · 2026-09-29NLPG：用于自进化语言智能体的自然语言策略梯度
NLPG: Natural-Language Policy Gradients for Self-Evolving Language Agents
arXiv 2609.33379 · 2026-09-29递归驾驭蒸馏：面向机器人操作的多智能体知识积累
Recursive Harness Distillation across Agents for Robot Manipulation
arXiv 2609.33378 · 2026-09-29结构化预测不确定性的最优传输Dropout
Optimal Transport Dropout for Structured Predictive Uncertainty
arXiv 2609.33377 · 2026-09-29基于TNF的谱嵌入在汽车保险欺诈检测中有效应用监督机器学习技术
TNF based Spectral Embedding for Effective Application of Supervised Machine Learning Techniques in Automobile Insurance Fraud Detection
arXiv 2609.33376 · 2026-09-29编解码器转换中存留的信息：用于语音深度伪造检测的池化无伴奏残差
What Survives the Codec Shift: Pooled No-Vocals Residuals for Speech Deepfake Detection
arXiv 2609.33375 · 2026-09-29具有平行挠率和常全纯截面曲率的Hermitian联络
Hermitian connections with parallel torsion and constant holomorphic sectional curvature
arXiv 2609.33374 · 2026-09-29身份辅助的无序DOA估计关联用于神经语音源跟踪
Identity-Assisted Association of Unordered DOA Estimates for Neural Speech Source Tracking
arXiv 2609.33373 · 2026-09-29从边际分布识别半鞅的可预测漂移
Identifying the Predictable Drift of a Semimartingale from Marginal Laws
arXiv 2609.33372 · 2026-09-29API 密钥绝不应成为 LLM 词汇表中的令牌：LLM 智能体系统中 API 凭据处理威胁分析及保险库中介执行边界的实证评估
API Secrets Should Never Become Tokens in the LLM's Vocabulary: A Threat Analysis of API Credential Handling in LLM Agent Systems and an Empirical Evaluation of a Vault-Mediated Execution Boundary
arXiv 2609.33371 · 2026-09-29选择规则决定胜负：开放集图异常检测的预注册审计
The Selection Rule Decides the Winner: A Pre-Registered Audit of Open-Set Graph Anomaly Detection
arXiv 2609.33370 · 2026-09-29几何感知多无人机全双工通信：系统设计与实验
Geometry-Aware Multi-UAV Full-Duplex Communication: System Design and Experiment
arXiv 2609.33369 · 2026-09-29DrafTS：面向时间序列建模的时间感知分解与残差校正
DrafTS: Time-Aware Decomposition with Residual Correction for Time Series Modeling
arXiv 2609.33368 · 2026-09-29迈向意识的全栈功能主义理论：识别其功能特征
Towards a full-stack functionalist theory of consciousness: Identifying its functional profile
arXiv 2609.33366 · 2026-09-29有限域上点到平面关联的组合方法
A combinatorial approach to point-flat incidences over finite fields
arXiv 2609.33365 · 2026-09-29银河系质子成分延伸至宇宙射线膝部之外的证据
Evidence for an extended Galactic proton component beyond the cosmic-ray knee
arXiv 2609.33364 · 2026-09-29几何自适应机制用于私有合成数据
Geometry-Adaptive Mechanisms for Private Synthetic Data
arXiv 2609.33363 · 2026-09-29从剧本到戏剧：一个用于可控多说话人对话语音合成的智能体框架
From Script to Drama: An Agentic Framework for Controllable Multi-Speaker Dialogue TTS
arXiv 2609.33362 · 2026-09-29伊辛自旋系统中的无温度临界性
Criticality without Temperature in an Ising Spin System
arXiv 2609.33361 · 2026-09-29加权初始条件下函数值次优性的精确预言机复杂度
Exact oracle complexity for function-value suboptimality under weighted initial conditions
arXiv 2609.33360 · 2026-09-29几何视图合成何时有助于酒标检索？一个跨自监督与视觉-语言骨干网络的公开单样本基准
When Does Geometric View Synthesis Help Wine Label Retrieval? A Public One-Shot Benchmark Across Self-Supervised and Vision-Language Backbones
arXiv 2609.33359 · 2026-09-29具有线性恢复力的Boussinesq方程的适定性
Well-posedness of Boussinesq Equation with Linear Restoring Force
arXiv 2609.33358 · 2026-09-29DISCERN：AI智能体能否像科学家一样工作并引导发现？
DISCERN: Can AI Agents Work Like Scientists and Guide Discovery?
arXiv 2609.33357 · 2026-09-29长时程模拟设计基准：在数小时模拟与混合信号电路设计任务上对智能体进行基准测试
Long-Horizon Analog Design Bench: Benchmarking Agents on Hours-Long Analog and Mixed-Signal Circuit Design Tasks
arXiv 2609.33356 · 2026-09-29揭开状态的面纱：状态自适应何时对掩码扩散语言模型重要
Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language Models
arXiv 2609.33355 · 2026-09-29可追踪的人到人形机器人手语基准测试
Traceable Human-to-Humanoid Sign Language Benchmarking
arXiv 2609.33354 · 2026-09-29聚焦与补充：双增强视觉Transformer用于多类异常分类
Focus and Supplement: Dual-Enhanced Vision Transformer for Multi-Class Anomaly Classification
arXiv 2609.33353 · 2026-09-29我的分类器中多少不精确性才是足够的不精确性？一种实用的引导程序
How Much Imprecision is Enough Imprecision in my Classifier? A Practical Elicitation Procedure
arXiv 2609.33352 · 2026-09-29QuPID：面向医学RAG的量子参数高效输入相关检索适配
QuPID: Quantum Parameter-Efficient Input-Dependent Retrieval Adaptation for Medical RAG
arXiv 2609.33351 · 2026-09-29KoopCell：基于Koopman的生成模型，用于从分布快照学习单细胞动力学
KoopCell: Koopman-Based Generative Model for Learning Single-Cell Dynamics from Distribution Snapshots
arXiv 2609.33350 · 2026-09-29面向含异常批次异构流数据的可再生在线期望回归
Renewable Online Expectile Regression for Heterogeneous Streaming Data with Abnormal Batches
arXiv 2609.33349 · 2026-09-29边界容许水平下 $\mathfrak{sl}_2$ 仿射顶点代数的经典自由性
Classical freeness of affine vertex algebras of $\mathfrak{sl}_2$ at boundary admissible levels
arXiv 2609.33348 · 2026-09-29MultiEcho：学习世界的实验科学
MultiEcho: An Experimental Science of Learned Worlds
arXiv 2609.33347 · 2026-09-29可测量才可管理：符号感知推荐需要符号感知评估
What Gets Measured Gets Managed: Sign-aware Recommendation Needs Sign-aware Evaluation
arXiv 2609.33346 · 2026-09-29语言判别提升多语言语音模型的语言学习能力
Language Discrimination Improves Linguistic Learning in Multilingual Speech Models
arXiv 2609.33345 · 2026-09-29
