2026-08论文中文摘要
半随机图过程中的三角饱和图
Triangle-Saturated Graphs in the Semi-Random Graph Process
arXiv 2608.14687 · 2026-08-18复向量三维场中的微分几何导论
Introduction to Differential Geometry in the Three-Dimensional Field of Complex Vectors
arXiv 2608.14686 · 2026-08-18一个分数，两个决策：针对罕见疾病尾部的选择性预测
One Score, Two Decisions: Selective Prediction on the Rare-Disease Tail
arXiv 2608.14683 · 2026-08-18RouteTS：面向时间序列预测的频时路由方法
RouteTS: Frequency-Time Routing for Time Series Forecasting
arXiv 2608.14682 · 2026-08-18自动还是受控？重复启动揭示基础大语言模型、指令调优大语言模型与人类的加工差异
Automatic or Controlled? Repetition Priming Reveals Divergent Processing in Base LLMs, Instruct LLMs, and Humans
arXiv 2608.14681 · 2026-08-18当智能体执行失败时：从遥测数据中检测和定位运行时故障
When Agentic Executions Fail: Detecting and Localizing Runtime Faults from Telemetry
arXiv 2608.14680 · 2026-08-18半导体工艺动力学的信息论因果建模
Information-Theoretic Causal Modelling of Semiconductor Process Dynamics
arXiv 2608.14678 · 2026-08-18离线环境光控制潜在扩散：架构、遥测与设备端评估
Offline Ambient-Controlled Latent Diffusion: Architecture, Telemetry, and On-Device Evaluation
arXiv 2608.14677 · 2026-08-18面向实时5G/6G信道估计的相位感知CNN：硬件在环验证
Phase-Aware CNN for Real-Time 5G/6G Channel Estimation with Hardware-in-the-loop Validation
arXiv 2608.14676 · 2026-08-18个性化考量：通用自监督学习表征在真实生活光体积描记（PPG）情感检测中的局限性
Take it Personally: The Limits of General SSL Representations for Real-Life PPG Emotion Detection
arXiv 2608.14675 · 2026-08-18基于导频与未知数据载荷的OFDM分布式天线系统中计算高效的被动定位联合最大似然估计器
A Computationally Efficient Joint Maximum Likelihood Estimator for Passive Localization in OFDM Distributed Antenna Systems with Pilots and Unknown Data Payloads
arXiv 2608.14674 · 2026-08-18面向多种高空平台站（HAPS）天线阵列架构的上行链路MIMO性能分析
Uplink MIMO Performance Analysis for Diverse HAPS Antenna Array Architectures
arXiv 2608.14672 · 2026-08-18迷人空间中D空间问题的密核与闭核归约
Dense-Kernel and Closed-Core Reductions in the D-space Problem for Charming Spaces
arXiv 2608.14671 · 2026-08-18ARGUS：用于利用Wi-Fi遥测技术实现可扩展人员识别的注意力引导Transformer
ARGUS: Attention-Guided Transformers for Scalable Person Identification Using Wi-Fi Telemetry
arXiv 2608.14670 · 2026-08-18超越正确性：基于 Lean 4 的自动化新颖性验证研究
Beyond Correctness: Toward Automated Novelty Verification with Lean 4
arXiv 2608.14669 · 2026-08-18BRA-Audit：基于累积暴露审计点放置的LLM多智能体系统预算运行时审计
BRA-Audit: Budgeted Runtime Auditing for LLM Multi-Agent Systems via Cumulative-Exposure Audit-Point Placement
arXiv 2608.14668 · 2026-08-18立场：科学团队中的智能体应作为人机系统（HAS）进行研究
Position: AI Agents in Scientific Teams Should Be Studied as Human-Agent Systems
arXiv 2608.14667 · 2026-08-18基于因果机制监测的跨域工业故障检测
Cross-Domain Industrial Fault Detection by Causal Mechanism Monitoring
arXiv 2608.14666 · 2026-08-18最佳采样温度何时随预算升高？Pass@k的充分条件
When Does the Best Sampling Temperature Rise with the Budget? Sufficient Conditions for Pass@k
arXiv 2608.14665 · 2026-08-18量化残差神经网络中的深度充分性：一阶准则
Quantifying Depth Sufficiency in Residual Neural Networks: A First-Order Criterion
arXiv 2608.14664 · 2026-08-18基于上下文学习评估建成环境对行动受限老年人感知社区步行便利性的影响
In-Context Learning to Assess Built Environment Impacts on Perceived Neighborhood Walkability Among Mobility-impaired Older Adults
arXiv 2608.14663 · 2026-08-18心脏能反映你的疼痛吗？用自监督心电表示学习应对X-ITE疼痛挑战赛
Does the Heart Show Your Pain? Tackling the X-ITE Pain Challenge with Self-Supervised ECG Representation Learning
arXiv 2608.14662 · 2026-08-18用于水文动态过程的时间推移电阻率层析成像反演的自动微分框架
An automatic-differentiation framework for time-lapse electrical resistivity tomography inversion of hydrologic dynamics
arXiv 2608.14661 · 2026-08-18基于环的空间Transformer：学习建筑分布与行人流量之间的非线性空间相互作用
Ring-based Spatial Transformer: Learning Non-linear Spatial Interactions between Building Distribution and Pedestrian Flow
arXiv 2608.14660 · 2026-08-18当不确定性还不够时：代码生成中自校正的实证研究
When Uncertainty Isn't Enough: An Empirical Study of Self-Correction in Code Generation
arXiv 2608.14659 · 2026-08-18pico-type：一种150万参数的字节级多头内容分类器
pico-type: A 1.5M-Parameter Byte-Level Multi-Head Content Classifier
arXiv 2608.14658 · 2026-08-18LUNG-KGMM：用于肺癌发病预测的知识引导多模态学习
LUNG-KGMM: Knowledge-Guided Multimodal Learning for Lung Cancer Incidence Prediction
arXiv 2608.14657 · 2026-08-18P2E-VQ：基于离散 patch 检索的、用于 PPG 的 ECG 关联表示增强方法
P2E-VQ: ECG-linked representation augmentation for PPG via discrete patch retrieval
arXiv 2608.14656 · 2026-08-18通过模态子空间激活诊断并缓解全模态大语言模型（Omni-LLMs）中的感知-决策失配问题
Diagnosing and Mitigating Perception-Decision Misalignment in Omni-LLMs via Modality Subspace Activation
arXiv 2608.14655 · 2026-08-18FedImp：基于杂质加权增强联邦学习收敛性
FedImp: Enhancing Federated Learning Convergence with Impurity-Based Weighting
arXiv 2608.14654 · 2026-08-18不确定信号是否有用？对带有回滚机制的不确定感知解码的系统研究
Do Uncertainty Signals Help? A Systematic Study of Uncertainty-Aware Decoding with Rollback Mechanisms
arXiv 2608.14653 · 2026-08-18通过数据缩放推动高分辨率天气预报的极限
Pushing the Limits of High-Resolution Weather Forecasting through Data Scaling
arXiv 2608.14652 · 2026-08-18评估面向无障碍灾害援助的跨文本与音频模态多模态大语言模型
Evaluating Multimodal LLMs across Text and Audio Modalities for Accessible Disaster Assistance
arXiv 2608.14651 · 2026-08-18预测衍生的中到全世界模型级联的配对精确重置评估
Paired Exact-Reset Evaluation of a Prediction-Derived Medium-to-Full World-Model Cascade
arXiv 2608.14650 · 2026-08-18离散扩散语言模型是无需训练的多标签分类器
Discrete Diffusion Language Models Are Training-Free Multi-Label Classifiers
arXiv 2608.14649 · 2026-08-18停止以全精度索引：重新审视向量嵌入的聚类方法
Stop Indexing at Full Precision: Revisiting Clustering for Vector Embeddings
arXiv 2608.14648 · 2026-08-18SMOPD：面向脏历史多回合在线策略自蒸馏的选择性令牌熵掩码
SMOPD: Selective Token-Entropy Masking for Dirty-History Multi-Turn On-Policy Self-Distillation
arXiv 2608.14647 · 2026-08-18iFuzz-Meta：一种连接自上而下与自下而上知识融合的可解释模糊学习框架
iFuzz-Meta: An Interpretable Fuzzy Learning Framework Bridging Top-Down and Bottom-Up Knowledge Integration
arXiv 2608.14646 · 2026-08-18用于CO₂反演的高效神经网络基高分辨率辐射传输及在干涉测量中的应用
Efficient Neural-Network-Based High-Resolution Radiative Transfer for CO___ Retrieval, and Application to Interferometric Sensing
arXiv 2608.14645 · 2026-08-18DUET：基于同权重分歧的双教师在线策略蒸馏用于禁止合规性
DUET: Dual-Teacher On-Policy Distillation via Same-Weight Disagreement for Prohibition Compliance
arXiv 2608.14644 · 2026-08-18证明门控发布：无服务器数据湖屋的提交前内容完整性验证
Proof-Gated Publication: Verify-Before-Commit Content Integrity for Serverless Data-Mesh Lakehouses
arXiv 2608.14643 · 2026-08-18基于每回合分布的伦理强化学习智能体训练与评估
Training and Evaluating Ethical Reinforcement Learning Agents on Per-Episode Distributions
arXiv 2608.14642 · 2026-08-18BDIP-Net：用于双层材料性能预测的双交互图学习方法
BDIP-Net: Dual-Interaction Graph Learning for Property Prediction of Bilayer Materials
arXiv 2608.14640 · 2026-08-18文档提取的有效逐字段选择性风险控制：三种失效模式、一个有效性阶梯及条件化何时奏效
Valid Per-Field Selective Risk Control for Document Extraction: Three Failure Modes, a Validity Ladder, and When Conditioning Pays
arXiv 2608.14639 · 2026-08-18随机初始化自编码器：不动点与混沌边缘
Randomly initialized autoencoders: fixed points and edge-of-chaos
arXiv 2608.14638 · 2026-08-18基于FlowBD-E1的铁铬液流电池早期循环充电轨迹生成预测与全生命周期健康管理
Early Cycle Charge Trajectory Generative Prediction and Full Life Cycle Health Management of Iron-Chromium Flow Batteries Based on FlowBD-E1
arXiv 2608.14637 · 2026-08-18分数阶优化器与分形激活函数：神经网络多尺度优化的实证研究
Fractional Optimizers Meet Fractal Activation Functions: An Empirical Study of Multi-Scale Optimization in Neural Network
arXiv 2608.14636 · 2026-08-18元可塑性作为增量学习的自适应梯度预条件化
Metaplasticity as adaptive gradient preconditioning for incremental learning
arXiv 2608.14634 · 2026-08-18471:1类别不平衡下的Wolff-Parkinson-White检测：数据瓶颈的泄漏控制研究
Wolff-Parkinson-White Detection at 471:1 Class Imbalance: A Leakage-Controlled Study of the Data Bottleneck
arXiv 2608.14633 · 2026-08-18DeMTS：将去噪轨迹作为多元时间序列用于扩散语言模型的幻觉检测
DeMTS: Denoising Trajectories as Multivariate Time Series for Hallucination Detection in Diffusion Language Models
arXiv 2608.14632 · 2026-08-18
