2026-09论文中文摘要

• 连续Gromov--Hausdorff距离的修改

  Modifications of the Continuous Gromov--Hausdorff Distance

  arXiv 2609.22201 · 2026-09-22

• PII-TRACE：多轮LLM对话中上下文感知PII检测的基准

  PII-TRACE: A Benchmark for Context-Aware PII Detection in Multi-Turn LLM Conversations

  arXiv 2609.22200 · 2026-09-22

• 通过跨层共享神经专家提升Transformer中的参数利用率

  Improving Parameter Utilization by Sharing Neural Experts Across Layers in Transformers

  arXiv 2609.22199 · 2026-09-22

• 人工智能在在线评论中的作用

  The Role of AI in Online Reviews

  arXiv 2609.22198 · 2026-09-22

• 剖析层级推理模型：一项机制性研究

  Dissecting Hierarchical Reasoning Models: A Mechanistic Study

  arXiv 2609.22197 · 2026-09-22

• EvoRank：LLM引导的多目标学习排序流水线进化

  EvoRank: LLM-Guided Evolution of Multi-Objective Learning-to-Rank Pipelines

  arXiv 2609.22196 · 2026-09-22

• 情境身份测试：区分持久认知身份与角色模仿

  The Situated Identity Test: Distinguishing Persistent Cognitive Identity from Persona Imitation

  arXiv 2609.22195 · 2026-09-22

• 一点SFT，一点RL：强化学习何时帮助长周期广告智能体

  A Pinch of SFT, A Dash of RL: When Reinforcement Learning Helps Long-Horizon Advertising Agents

  arXiv 2609.22194 · 2026-09-22

• 量子稳定鲁棒主成分分析：理论与NISQ体制下的证据

  Quantum-Stable Robust Principal Component Analysis: Theory and Evidence from NISQ Regimes

  arXiv 2609.22193 · 2026-09-22

• SegTSim：面向异构多变量系统的基于大数据的分段时间模拟框架

  SegTSim: A Big Data Driven Segmented Temporal Simulation Framework for Heterogeneous Multivariate Systems

  arXiv 2609.22192 · 2026-09-22

• WildfireSpreadBench：在野火蔓延预测中，指标决定模型

  WildfireSpreadBench: The Metric Decides the Model in Wildfire Spread Prediction

  arXiv 2609.22191 · 2026-09-22

• 物理信息经典与量子神经网络用于一维薛定谔本征值问题

  Physics-Informed Classical and Quantum Neural Networks for One-Dimensional Schrodinger Eigenvalue Problems

  arXiv 2609.22189 · 2026-09-22

• 超越匿名化的公平性？德语大语言模型生成简历中的人口统计信息泄露

  Fairness Beyond Anonymization? Demographic Leakage in German LLM-Generated Resumes

  arXiv 2609.22188 · 2026-09-22

• 相关性引导的流匹配与退火掩码用于空间转录组生成

  Correlation-Guided Flow Matching with Annealed Masking for Spatial Transcriptomics Generation

  arXiv 2609.22187 · 2026-09-22

• 自适应物理信息神经网络用于Blasius边界层问题

  Adaptive Physics-Informed Neural Networks for the Blasius Boundary-Layer Problem

  arXiv 2609.22185 · 2026-09-22

• DPTM-DT：用于药物-靶点预测的双预训练Transformer多任务表示学习

  DPTM-DT: Dual-Pretrained Transformer Multitask Representation Learning for Drug-Target Prediction

  arXiv 2609.22184 · 2026-09-22

• CleanScore：带阴性对照与敏感性边界的黑盒基准审计

  CleanScore: Black-Box Benchmark Audits with Negative Controls and Sensitivity Bounds

  arXiv 2609.22183 · 2026-09-22

• 高斯过程去相关用于基于时空深度学习的雪水当量预测

  Gaussian Process Decorrelation for Spatiotemporal Deep Learning-Based Snow Water Equivalent Prediction

  arXiv 2609.22182 · 2026-09-22

• 可局域化的二分秩一理想测量具有块复制尼斯-贝尔结构

  Localizable Bipartite Rank-One Ideal Measurements Have a Block-Replicated Nice-Bell Structure

  arXiv 2609.22181 · 2026-09-22

• 静态多阈值采样的非均匀采样理论

  A Nonuniform-Sampling Theory for Static Multi-Threshold Sampling

  arXiv 2609.22180 · 2026-09-22

• 基于基序异常分析的心电信号可解释压力检测

  Interpretable Stress Detection from ECG Signals Using Motif-Based Anomaly Analysis

  arXiv 2609.22179 · 2026-09-22

• OpenBlock：自适应拼图匹配游戏的建设性与验证内容生成

  OpenBlock: Constructive and Verified Content Generation for Adaptive Tile-Matching Games

  arXiv 2609.22177 · 2026-09-22

• 六个双量子比特门的每种架构在三量子比特上都是局部通用的

  Every architecture of six two-qubit gates is locally universal on three qubits

  arXiv 2609.22176 · 2026-09-22

• 对比世界模型

  Contrastive World Models

  arXiv 2609.22175 · 2026-09-22

• SCoR：一个用于预测科学概念间关系的层级框架

  SCoR: A Hierarchical Framework for Forecasting Relations Between Scientific Concepts

  arXiv 2609.22174 · 2026-09-22

• 工业运动学轨迹模型（IKTM）：无坐标自回归生成器

  Industrial Kinematic Trajectory Model (IKTM): Coordinate-Free Autoregressive Generator

  arXiv 2609.22173 · 2026-09-22

• 基于小波和STFT特征的电驱动桥质量监测的分阶段异常检测

  Stagewise Anomaly Detection for E-Transaxle Quality Monitoring Using Wavelet and STFT Features

  arXiv 2609.22172 · 2026-09-22

• 量化隐藏盐分以实现精准医疗：基于联合因子检索与思维链推理的钠摄入评估

  Quantifying Hidden Salt for Precision Healthcare: Sodium Assessment via Joint-Factor Retrieval and Chain-of-Thought Inference

  arXiv 2609.22171 · 2026-09-22

• 多重潜在排序能更好地预测语言模型的偏好

  Multiple latent orderings better predict language model preferences

  arXiv 2609.22170 · 2026-09-22

• 单一文化偏见：大型语言模型中的相关偏见导致招聘中不平等的系统性排斥率

  Monocultural Biases: Correlated biases in large language models lead to unequal systemic exclusion rates in hiring

  arXiv 2609.22169 · 2026-09-22

• 一种用于监测老年人健康的节能可穿戴设备

  An Energy Efficient Wearable for Monitoring Elderly People Health

  arXiv 2609.22168 · 2026-09-22

• 面向反应产率预测的角色感知摩根指纹

  Role-Aware Morgan Fingerprints for Reaction Yield Prediction

  arXiv 2609.22167 · 2026-09-22

• 基于聚类的物联网系统集体异常检测：一种图神经网络方法

  Clustering-Based Collective Anomaly Detection in IoT Systems: A Graph Neural Network Approach

  arXiv 2609.22166 · 2026-09-22

• TARGet：基于图神经网络的拓扑感知融合射频电路功能建模

  TARGet: Topology-Aware Fusion-based Radio Frequency Circuit Functional Modeling using Graph Neural Networks

  arXiv 2609.22165 · 2026-09-22

• 多智能体流水线：从纵向结构化电子健康记录生成基于来源的合成临床笔记

  A Multi-Agent Pipeline for Source-Grounded Synthetic Note Generation from Longitudinal Structured EHR

  arXiv 2609.22164 · 2026-09-22

• MechaTerp-TRACE：语言模型组件消融分析的新方法

  MechaTerp-TRACE: A Novel Approach for Component Ablation Analysis in Language Models

  arXiv 2609.22163 · 2026-09-22

• 超越原始上下文传输：基于表示的联邦检索增强生成

  Beyond Raw Context Transfer: Representation-based Federated Retrieval-Augmented Generation

  arXiv 2609.22162 · 2026-09-22

• 说教性知识还是临床病例？数据类型如何塑造医学大语言模型

  Didactic knowledge or Clinical Cases? How Data Types Shape Medical Large Language Models

  arXiv 2609.22161 · 2026-09-22

• 半导体制造中不确定性与业务感知的剩余使用寿命估计

  Uncertainty and Business-Aware Remaining Useful Life Estimation for Semiconductor Manufacturing

  arXiv 2609.22160 · 2026-09-22

• 极限及其根据：QBism、科学实在论与玻恩规则

  The Limit and Its Ground. QBism, Scientific Realism, and the Born Rule

  arXiv 2609.22159 · 2026-09-22

• 投机极限：界定混合专家模型中的投机解码

  The Limits of Speculation: Bounding Speculative Decoding in Mixture-of-Experts

  arXiv 2609.22156 · 2026-09-22

• 从潜在生物标志物到临床规则：基于嵌入引导的规则挖掘与基于归因的翻译用于可解释表格学习

  From Latent Biomarkers to Clinical Rules: Embedding-Guided Rule Mining and Attribution-Based Translation for Interpretable Tabular Learning

  arXiv 2609.22155 · 2026-09-22

• 表格数据基础模型评估的比较框架：医疗保健案例研究

  A Comparative Framework for Evaluating Foundation Models on Tabular Data: A Case Study in Healthcare

  arXiv 2609.22154 · 2026-09-22

• SafeTune：一个用于审计和修复微调大语言模型中安全漂移的统一忠实库

  SafeTune: A Unified Faithful Library for Auditing and Repairing Safety Drift in Fine-Tuned LLMs

  arXiv 2609.22153 · 2026-09-22

• 想象力源自幻觉吗？大语言模型中想象力与幻觉的跨分类学评估

  Is Imagination Derived from Hallucination? A Cross-Taxonomy Evaluation of Imagination and Hallucination in Large Language Models

  arXiv 2609.22152 · 2026-09-22

• 语言模型是否知道自身的约束？

  Do Language Models Know Their Own Constraints?

  arXiv 2609.22151 · 2026-09-22

• ECP-Bench：使用基础模型进行娱乐内容推广的基准测试与学习

  ECP-Bench: Benchmarking and Learning Entertainment Content Promotion with Foundation Models

  arXiv 2609.22150 · 2026-09-22

• 超越拼接假设：基于语义量化的多模态合成数据评估统一框架

  Beyond the Stitching Assumption: A Unified Framework for Multimodal Synthetic Data Evaluation via Semantic Quantization

  arXiv 2609.22149 · 2026-09-22

• SleepEffFormer：用于单通道脑电睡眠分期的高效CNN-Transformer与转换感知平滑

  SleepEffFormer: Efficient CNN-Transformer with Transition-Aware Smoothing for Single-Channel EEG Sleep Stage Classification

  arXiv 2609.22148 · 2026-09-22

• WaveFoG：用于帕金森冻结步态检测的小波门控Transformer（基于可穿戴加速度计信号）

  WaveFoG: Wavelet Gated Transformer for Parkinson's Freezing of Gait Detection Using Wearable Accelerometer Signals

  arXiv 2609.22147 · 2026-09-22