机器学习学术速递[9.3]

点击阅读原文访问arxivdaily.com，涵盖CS|物理|数学|经济|统计|金融|生物|电气领域，更有搜索、收藏等功能！

cs.LG 方向，今日共计380篇

大模型相关(36篇)

【1】Jointly Reinforcing Diversity and Quality in Language Model Generations
标题：共同增强语言模型生成的多样性和质量
链接：https://arxiv.org/abs/2509.02534

【2】Comparative Study of Pre-Trained BERT and Large Language Models for Code-Mixed Named Entity Recognition
标题：预训练BERT和大语言模型用于混合代码命名实体识别的比较研究
链接：https://arxiv.org/abs/2509.02514

【3】MLP-Offload: Multi-Level, Multi-Path Offloading for LLM Pre-training to Break the GPU Memory Wall
标题：MLP卸载：LLM预训练的多级别、多路径卸载，以打破图形处理器内存墙
链接：https://arxiv.org/abs/2509.02480

【4】Do LLMs Adhere to Label Definitions? Examining Their Receptivity to External Label Definitions
标题：LLM遵守标签定义吗？检查其对外部标签定义的接受性
链接：https://arxiv.org/abs/2509.02452

【5】An Ensemble Classification Approach in A Multi-Layered Large Language Model Framework for Disease Prediction
标题：用于疾病预测的多层大型语言模型框架中的整体分类方法
链接：https://arxiv.org/abs/2509.02446

【6】Cache Management for Mixture-of-Experts LLMs -- extended version
标题：专家混合LLM的缓存管理-扩展版本
链接：https://arxiv.org/abs/2509.02408

【7】Scale, Don't Fine-tune: Guiding Multimodal LLMs for Efficient Visual Place Recognition at Test-Time
标题：规模，不要微调：指导多模式LLM在测试时实现高效的视觉位置识别
链接：https://arxiv.org/abs/2509.02129

【8】When LLM Meets Time Series: Can LLMs Perform Multi-Step Time Series Reasoning and Inference
标题：当LLM遇到时间序列时：LLM能否执行多步时间序列推理和推理
链接：https://arxiv.org/abs/2509.01822

【9】Flaw or Artifact? Rethinking Prompt Sensitivity in Evaluating LLMs
标题：缺陷还是毛刺？重新思考评估LLM的即时敏感性
链接：https://arxiv.org/abs/2509.01790

【10】Communication-Aware Knowledge Distillation for Federated LLM Fine-Tuning over Wireless Networks
标题：无线网络上的联邦LLM微调的通信感知知识提炼
链接：https://arxiv.org/abs/2509.01750

【11】Benchmarking Optimizers for Large Language Model Pretraining
标题：大型语言模型预训练的基准优化器
链接：https://arxiv.org/abs/2509.01440

【12】DPF-CM: A Data Processing Framework with Privacy-Preserving Vector Databases for Chinese Medical LLMs Training and Deployment
标题：DPF-CM：用于中国医学LLM训练和部署的具有隐私保护载体数据库的数据处理框架
链接：https://arxiv.org/abs/2509.01354

【13】Iterative In-Context Learning to Enhance LLMs Abstract Reasoning: The Case-Study of Algebraic Tasks
标题：迭代上下文学习增强LLM抽象推理：代数任务的案例研究
链接：https://arxiv.org/abs/2509.01267

【14】LiquidGEMM: Hardware-Efficient W4A8 GEMM Kernel for High-Performance LLM Serving
标题：LiquidGEMM：硬件高效的W4 A8 GEMM内核，用于高性能LLM服务
链接：https://arxiv.org/abs/2509.01229

【15】DaMoC: Efficiently Selecting the Optimal Large Language Model for Fine-tuning Domain Taks Based on Data and Model Compression
标题：DaMoC：基于数据和模型压缩有效选择最佳大型语言模型进行微调领域获取
链接：https://arxiv.org/abs/2509.01221

【16】Do Video Language Models Really Know Where to Look? Diagnosing Attention Failures in Video Language Models
标题：视频语言模型真的知道去哪里看吗？诊断视频语言模型中的注意力故障
链接：https://arxiv.org/abs/2509.01167

【17】Analysis of Error Sources in LLM-based Hypothesis Search for Few-Shot Rule Induction
标题 ：基于LLM的Few-Shot规则归纳假设搜索中的误差源分析
链接：https://arxiv.org/abs/2509.01016

【18】Self-Exploring Language Models for Explainable Link Forecasting on Temporal Graphs via Reinforcement Learning
标题：基于强化学习的时态图可解释链接预测的自探索语言模型
链接：https://arxiv.org/abs/2509.00975

【19】SATQuest: A Verifier for Logical Reasoning Evaluation and Reinforcement Fine-Tuning of LLMs
标题：SATQuest：LLM逻辑推理评估与强化微调验证器
链接：https://arxiv.org/abs/2509.00930

【20】CaresAI at BioCreative IX Track 1 -- LLM for Biomedical QA
标题：CaresAI在BioCreative IX Track 1 --生物医学QA法学硕士
链接：https://arxiv.org/abs/2509.00806

【21】Efficient Graph Understanding with LLMs via Structured Context Injection
标题：基于结构化上下文注入的LLM的高效图理解
链接：https://arxiv.org/abs/2509.00740

【22】Universal Properties of Activation Sparsity in Modern Large Language Models
标题：现代大型语言模型中激活稀疏性的普遍性质
链接：https://arxiv.org/abs/2509.00454

【23】Metis: Training Large Language Models with Advanced Low-Bit Quantization
标题：Metis：利用高级低位量化训练大型语言模型
链接：https://arxiv.org/abs/2509.00404

【24】The Resurgence of GCG Adversarial Attacks on Large Language Models
标题：GCG对大型语言模型的对抗性攻击的卷土重来
链接：https://arxiv.org/abs/2509.00391

【25】LLM-Driven Policy Diffusion: Enhancing Generalization in Offline Reinforcement Learning
标题：法学硕士驱动的政策传播：增强离线强化学习的概括性
链接：https://arxiv.org/abs/2509.00347

【26】Mechanistic interpretability for steering vision-language-action models
标题：引导视觉-语言-动作模型的机械可解释性
链接：https://arxiv.org/abs/2509.00328

【27】Learning to Shard: RL for Co-optimizing the Parallelism Degrees and Per-operator Sharding Dimensions in Distributed LLM Inference
标题：学习碎片：RL用于在分布式LLM推理中协同优化并行度和每个操作员碎片维度
链接：https://arxiv.org/abs/2509.00217

【28】Pre-trained knowledge elevates large language models beyond traditional chemical reaction optimizers
标题：预先训练的知识将大型语言模型提升到传统化学反应优化器之外
链接：https://arxiv.org/abs/2509.00103

【29】LLM-QUBO: An End-to-End Framework for Automated QUBO Transformation from Natural Language Problem Descriptions
标题：LLM-QUBO：从自然语言问题描述自动进行QUBO转换的端到端框架
链接：https://arxiv.org/abs/2509.00099

【30】Pruning Weights but Not Truth: Safeguarding Truthfulness While Pruning LLMs
标题：修剪权重而不是真相：在修剪LLM的同时保护真实性
链接：https://arxiv.org/abs/2509.00096

【31】Learning to Refine: Self-Refinement of Parallel Reasoning in LLMs
标题：学习完善：LLM中并行推理的自我完善
链接：https://arxiv.org/abs/2509.00084

【32】Language and Experience: A Computational Model of Social Learning in Complex Tasks
标题：语言与经验：复杂任务中社会学习的计算模型
链接：https://arxiv.org/abs/2509.00074

【33】AnomalyExplainer Explainable AI for LLM-based anomaly detection using BERTViz and Captum
标题：AnomalyExplainer使用BERTViz和Captum进行基于LLM的异常检测的可解释人工智能
链接：https://arxiv.org/abs/2509.00069

【34】Exploring and Reshaping the Weight Distribution in LLM
标题：探索和重塑LLM中的权重分布
链接：https://arxiv.org/abs/2509.00046

【35】Diagnosing Psychiatric Patients: Can Large Language and Machine Learning Models Perform Effectively in Emergency Cases?
标题：诊断精神病患者：大型语言和机器学习模型能否在紧急情况下有效执行？
链接：https://arxiv.org/abs/2509.00026

【36】NoLBERT: A No Lookahead(back) Foundational Language Model for Empirical Research
标题：NoLBERT：一个不向前看（向后）的实证研究基础语言模型
链接：https://arxiv.org/abs/2509.01110

Graph相关(图学习|图神经网络|图优化等)(22篇)

【1】HydroGAT: Distributed Heterogeneous Graph Attention Transformer for Spatiotemporal Flood Prediction
标题：HydroGAT：用于时空洪水预测的分布式异类图注意力Transformer
链接：https://arxiv.org/abs/2509.02481

【2】Exploring Variational Graph Autoencoders for Distribution Grid Data Generation
标题：探索用于配电网数据生成的变分图自动编码器
链接：https://arxiv.org/abs/2509.02469

【3】HiGraph: A Large-Scale Hierarchical Graph Dataset for Malware Analysis
标题：HiGraph：一个用于恶意软件分析的大规模层次图数据集
链接：https://arxiv.org/abs/2509.02113

【4】Second-Order Tensorial Partial Differential Equations on Graphs
标题：图上的二阶张量偏方程
链接：https://arxiv.org/abs/2509.02015

【5】ACA-Net: Future Graph Learning for Logistical Demand-Supply Forecasting
标题：ACA-Net：物流供需预测的未来图学习
链接：https://arxiv.org/abs/2509.01997

【6】TransGAT: Transformer-Based Graph Neural Networks for Multi-Dimensional Automated Essay Scoring
标题：TransGAT：基于转换器的图神经网络用于多维自动论文评分
链接：https://arxiv.org/abs/2509.01640

【7】Ultra Fast Warm Start Solution for Graph Recommendations
标题：图形推荐的超快速热启动解决方案
链接：https://arxiv.org/abs/2509.01549

【8】Graph Contrastive Learning versus Untrained Baselines: The Role of Dataset Size
标题：图表对比学习与未经训练的基线：数据集大小的作用
链接：https://arxiv.org/abs/2509.01541

【9】Learn to Jump: Adaptive Random Walks for Long-Range Propagation through Graph Hierarchies
标题：学习跳跃：通过图层次进行远程传播的自适应随机游走
链接：https://arxiv.org/abs/2509.01381

【10】MatPROV: A Provenance Graph Dataset of Material Synthesis Extracted from Scientific Literature
标题：MatPROV：从科学文献中提取的材料合成源图数据集
链接：https://arxiv.org/abs/2509.01042

【11】Superposition in Graph Neural Networks
标题：图神经网络中的叠加
链接：https://arxiv.org/abs/2509.00928

【12】Flow Matters: Directional and Expressive GNNs for Heterophilic Graphs
标题：Flow Matters：异嗜性图的方向性和表达性GNN
链接：https://arxiv.org/abs/2509.00772

【13】Task-Aware Adaptive Modulation: A Replay-Free and Resource-Efficient Approach For Continual Graph Learning
标题：任务感知自适应调制：连续图形学习的免回放且资源高效的方法
链接：https://arxiv.org/abs/2509.00735

【14】RoFt-Mol: Benchmarking Robust Fine-Tuning with Molecular Graph Foundation Models
标题：RoFt-Mol：利用分子图基础模型进行稳健微调基准
链接：https://arxiv.org/abs/2509.00614

【15】Biological Pathway Informed Models with Graph Attention Networks (GATs)
标题：具有图形注意力网络（GAT）的生物路径知情模型
链接：https://arxiv.org/abs/2509.00524

【16】Graph Convolutional Network With Pattern-Spatial Interactive and Regional Awareness for Traffic Forecasting
标题：具有模式空间交互和区域感知的交通预测图卷积网络
链接：https://arxiv.org/abs/2509.00515

【17】Unifying Adversarial Perturbation for Graph Neural Networks
标题：图神经网络的统一对抗扰动
链接：https://arxiv.org/abs/2509.00387

【18】Design of Experiment for Discovering Directed Mixed Graph
标题：发现有向混合图的实验设计
链接：https://arxiv.org/abs/2509.01887

【19】Reinforcement learning for graph theory, Parallelizing Wagner's approach
标题：用于图形理论的强化学习，使瓦格纳的方法平行化
链接：https://arxiv.org/abs/2509.01607

【20】Enabling Down Syndrome Research through a Knowledge Graph-Driven Analytical Framework
标题：通过知识图驱动的分析框架实现唐氏综合症研究
链接：https://arxiv.org/abs/2509.01565

【21】Hybrid Topic-Semantic Labeling and Graph Embeddings for Unsupervised Legal Document Clustering
标题：无监督法律文档集群的混合主题-语义标记和图嵌入
链接：https://arxiv.org/abs/2509.00990

【22】Deep Learning for Operational High-Resolution Nowcasting in Switzerland Using Graph Neural Networks
标题：使用图神经网络在瑞士进行深度学习用于运营高分辨率即时预报
链接：https://arxiv.org/abs/2509.00017

Transformer(15篇)

【1】Generative Sequential Notification Optimization via Multi-Objective Decision Transformers
标题：通过多目标决策转换器进行生成式顺序通知优化
链接：https://arxiv.org/abs/2509.02458

【2】Speech transformer models for extracting information from baby cries
标题：用于从婴儿哭声中提取信息的语音Transformer模型
链接：https://arxiv.org/abs/2509.02259

【3】From Attack Descriptions to Vulnerabilities: A Sentence Transformer-Based Approach
标题：从攻击描述到漏洞：基于句子转换器的方法
链接：https://arxiv.org/abs/2509.02077

【4】GradES: Significantly Faster Training in Transformers with Gradient-Based Early Stopping
标题：GradES：通过基于学生的早期停止，Transformer的训练速度明显加快
链接：https://arxiv.org/abs/2509.01842

【5】A Multi-target Bayesian Transformer Framework for Predicting Cardiovascular Disease Biomarkers during Pandemics
标题：用于预测大流行期间心血管疾病生物标志物的多目标Bayesian Transformer框架
链接：https://arxiv.org/abs/2509.01794

【6】Learning to Ask: Decision Transformers for Adaptive Quantitative Group Testing
标题：学会提问：自适应量化群体测试的决策变革者
链接：https://arxiv.org/abs/2509.01723

【7】Efficient Transformer-Inspired Variants of Physics-Informed Deep Operator Networks
标题：受物理知识启发的深度运营商网络的高效变形
链接：https://arxiv.org/abs/2509.01679

【8】SCOUT: Toward Sub-Quadratic Attention via Segment Compression for Optimized Utility in Transformers
标题：SCOUT：通过分段压缩实现次二次注意力，以优化Transformer的效用
链接：https://arxiv.org/abs/2509.00935

【9】DTRNet: Dynamic Token Routing Network to Reduce Quadratic Costs in Transformers
标题：DTRNet：动态令牌路由网络，降低Transformer中的二次成本
链接：https://arxiv.org/abs/2509.00925

【10】Forecasting the Ionosphere from Sparse GNSS Data with Temporal-Fusion Transformers
标题：利用时间融合变换器从稀疏的GNSS数据预测电离层
链接：https://arxiv.org/abs/2509.00631

【11】Memory Limitations of Prompt Tuning in Transformers
标题：Transformer中即时调谐的记忆限制
链接：https://arxiv.org/abs/2509.00421

【12】Evaluating the Effectiveness of Transformer Layers in Wav2Vec 2.0, XLS-R, and Whisper for Speaker Identification Tasks
标题：评估Wav 2 Vec 2.0、XLS-R和Whisper中Transformer层对说话人识别任务的有效性
链接：https://arxiv.org/abs/2509.00230

【13】From TLinFormer to TConstFormer: The Leap to Constant-Time Transformer Attention: Achieving O(1) Computation and O(1) KV Cache during Autoregressive Inference
标题：从TLinFormer到TConstFormer：注意：在自回归推理中实现O（1）计算和O（1）KV缓存
链接：https://arxiv.org/abs/2509.00202

【14】Scaling Legal AI: Benchmarking Mamba and Transformers for Statutory Classification and Case Law Retrieval
标题：扩展法律人工智能：对Mamba和Transformers进行统计分类和案例法检索的基准测试
链接：https://arxiv.org/abs/2509.00141

【15】Resting-state fMRI Analysis using Quantum Time-series Transformer
标题：使用量子时间序列Transformer的静息状态fMRI分析
链接：https://arxiv.org/abs/2509.00711

GAN|对抗|攻击|生成相关(19篇)

【1】Unifi3D: A Study on 3D Representations for Generation and Reconstruction in a Common Framework
标题：Unifi 3D：在通用框架中生成和重建的3D表示研究
链接：https://arxiv.org/abs/2509.02474

【2】Conditional-$t^3$VAE: Equitable Latent Space Allocation for Fair Generation
标题：有条件的-$t$VAE：公平发电的公平潜在空间分配
链接：https://arxiv.org/abs/2509.02154

【3】Abex-rat: Synergizing Abstractive Augmentation and Adversarial Training for Classification of Occupational Accident Reports
标题：Abex-rat：协同抽象增强和对抗性训练用于职业事故报告分类
链接：https://arxiv.org/abs/2509.02072

【4】Knowledge distillation as a pathway toward next-generation intelligent ecohydrological modeling systems
标题：知识提炼作为下一代智能生态水文建模系统的途径
链接：https://arxiv.org/abs/2509.01972

【5】Prediction, Generation of WWTPs microbiome community structures and Clustering of WWTPs various feature attributes using DE-BP model, SiTime-GAN model and DPNG-EPMC ensemble clustering algorithm with modulation of microbial ecosystem health
标题：预测、WWTP微生物组群落结构的生成以及使用DE-BP模型、SiTime-GAN模型和DPNG-EPMC集成集群算法对WWTP各种特征属性进行集群，并调节微生物生态系统健康
链接：https://arxiv.org/abs/2509.01526

【6】Geometric origin of adversarial vulnerability in deep learning
标题：深度学习中对抗脆弱性的几何起源
链接：https://arxiv.org/abs/2509.01235

【7】RAMS: Residual-based adversarial-gradient moving sample method for scientific machine learning in solving partial differential equations
标题：RAMS：基于剩余的对抗梯度移动样本方法，用于求解偏微方程的科学机器学习
链接：https://arxiv.org/abs/2509.01234

【8】Sequential Difference Maximization: Generating Adversarial Examples via Multi-Stage Optimization
标题：序列差异最大化：通过多阶段优化生成对抗性示例
链接：https://arxiv.org/abs/2509.00826

【9】ProCause: Generating Counterfactual Outcomes to Evaluate Prescriptive Process Monitoring Methods
标题：ProCause：生成反事实结果以评估规定过程监控方法
链接：https://arxiv.org/abs/2509.00797

【10】The Name-Free Gap: Policy-Aware Stylistic Control in Music Generation
标题：无名差距：音乐世代中的政策意识风格控制
链接：https://arxiv.org/abs/2509.00654

【11】Learning from Peers: Collaborative Ensemble Adversarial Training
标题：向同行学习：协作整体对抗训练
链接：https://arxiv.org/abs/2509.00089

【12】Entropy-Guided Loop: Achieving Reasoning through Uncertainty-Aware Generation
标题：熵引导循环：通过不确定性感知生成实现推理
链接：https://arxiv.org/abs/2509.00079

【13】SynCircuit: Automated Generation of New Synthetic RTL Circuits Can Enable Big Data in Circuits
标题：SynCircuit：自动生成新的合成RTL电路可以在电路中实现大数据
链接：https://arxiv.org/abs/2509.00071

【14】Scaffold Diffusion: Sparse Multi-Category Voxel Structure Generation with Discrete Diffusion
标题：支架扩散：利用离散扩散生成稀疏多类别体素结构
链接：https://arxiv.org/abs/2509.00062

【15】Mitigating Data Exfiltration Attacks through Layer-Wise Learning Rate Decay Fine-Tuning
标题：通过分层学习率衰减微调缓解数据泄露攻击
链接：https://arxiv.org/abs/2509.00027

【16】Amputation-imputation based generation of synthetic tabular data for ratemaking
标题：基于截肢-插补的合成表格数据生成用于费率制定
链接：https://arxiv.org/abs/2509.02171

【17】Non-Identical Diffusion Models in MIMO-OFDM Channel Generation
标题：MMO-CDMA通道生成中的非相同扩散模型
链接：https://arxiv.org/abs/2509.01641

【18】Semi-Supervised Bayesian GANs with Log-Signatures for Uncertainty-Aware Credit Card Fraud Detection
标题：具有日志签名的半监督Bayesian GAN用于不确定性信用卡欺诈检测
链接：https://arxiv.org/abs/2509.00931

【19】Conditional Generative Adversarial Networks Based Inertial Signal Translation
标题：基于条件生成对抗网络的惯性信号翻译
链接：https://arxiv.org/abs/2509.00016

半/弱/无/有监督|不确定性|主动学习(15篇)

【1】Implicit Actor Critic Coupling via a Supervised Learning Framework for RLVR
标题：通过RLVR监督学习框架的隐性演员批评者耦合
链接：https://arxiv.org/abs/2509.02522

【2】EmoPerso: Enhancing Personality Detection with Self-Supervised Emotion-Aware Modelling
标题：Perso：通过自我监督的描述感知建模增强人格检测
链接：https://arxiv.org/abs/2509.02450

【3】Online Identification of IT Systems through Active Causal Learning
标题：通过主动因果学习在线识别IT系统
链接：https://arxiv.org/abs/2509.02130

【4】Learning Longitudinal Stress Dynamics from Irregular Self-Reports via Time Embeddings
标题：通过时间嵌入从不规则的自我报告中学习纵向压力动态
链接：https://arxiv.org/abs/2509.01569

【5】Unified Supervision For Vision-Language Modeling in 3D Computed Tomography
标题：3D计算机断层扫描中视觉语言建模的统一监督
链接：https://arxiv.org/abs/2509.01554

【6】Unsupervised Identification and Replay-based Detection (UIRD) for New Category Anomaly Detection in ECG Signal
标题：心电信号新类别异常检测的无监督识别和回放检测（UIRD）
链接：https://arxiv.org/abs/2509.01512

【7】M3Ret: Unleashing Zero-shot Multimodal Medical Image Retrieval via Self-Supervision
标题：M3 Ret：通过自我监督释放Zero-Shot多模式医学图像检索
链接：https://arxiv.org/abs/2509.01360

【8】Towards Trustworthy Vital Sign Forecasting: Leveraging Uncertainty for Prediction Intervals
标题：迈向值得信赖的重要迹象预测：利用不确定性来预测间隔
链接：https://arxiv.org/abs/2509.01319

【9】Ultra Strong Machine Learning: Teaching Humans Active Learning Strategies via Automated AI Explanations
标题：超强机器学习：通过自动化人工智能简化教授人类主动学习策略
链接：https://arxiv.org/abs/2509.00961

【10】Predicting Multi-Type Talented Students in Secondary School Using Semi-Supervised Machine Learning
标题：利用半监督机器学习预测中学多类型优秀学生
链接：https://arxiv.org/abs/2509.00863

【11】Why Pool When You Can Flow? Active Learning with GFlowNets
标题：当你可以流动时为什么要进行池？使用GFlowNets进行主动学习
链接：https://arxiv.org/abs/2509.00704

【12】FedThief: Harming Others to Benefit Oneself in Self-Centered Federated Learning
标题：FedThief：在以自我为中心的联邦学习中伤害他人以造福自己
链接：https://arxiv.org/abs/2509.00540

【13】Variational Uncertainty Decomposition for In-Context Learning
标题：上下文内学习的变分不确定性分解
链接：https://arxiv.org/abs/2509.02327

【14】Wrong Model, Right Uncertainty: Spatial Associations for Discrete Data with Misspecification
标题：错误的模型，正确的不确定性：具有错误规范的离散数据的空间关联
链接：https://arxiv.org/abs/2509.01776

【15】Self-Organising Memristive Networks as Physical Learning Systems
标题：自组织记忆网络作为物理学习系统
链接：https://arxiv.org/abs/2509.00747

迁移|Zero/Few/One-Shot|自适应(14篇)

【1】Federated learning over physical channels: adaptive algorithms with near-optimal guarantees
标题：物理通道上的联合学习：具有接近最优保证的自适应算法
链接：https://arxiv.org/abs/2509.02538

【2】Ordinal Adaptive Correction: A Data-Centric Approach to Ordinal Image Classification with Noisy Labels
标题：有序自适应纠正：一种以数据为中心的具有噪音标签的有序图像分类方法
链接：https://arxiv.org/abs/2509.02351

【3】AdaSwitch: An Adaptive Switching Meta-Algorithm for Learning-Augmented Bounded-Influence Problems
标题：AdaSwitch：一种用于学习增强有界影响问题的自适应交换元算法
链接：https://arxiv.org/abs/2509.02302

【4】VISP: Volatility Informed Stochastic Projection for Adaptive Regularization
标题：VISP：适应性正规化的波动性知情随机预测
链接：https://arxiv.org/abs/2509.01903

【5】One-Shot Clustering for Federated Learning Under Clustering-Agnostic Assumption
标题：预测不可知假设下的联邦学习一次集群
链接：https://arxiv.org/abs/2509.01587

【6】ADMP-GNN: Adaptive Depth Message Passing GNN
标题：ADMP-GNN：自适应深度消息传递GNN
链接：https://arxiv.org/abs/2509.01170

【7】MATL-DC: A Multi-domain Aggregation Transfer Learning Framework for EEG Emotion Recognition with Domain-Class Prototype under Unseen Targets
标题：MATL-DC：一个用于脑电情感识别的多域聚集转移学习框架，具有不可见目标下的领域类原型
链接 ：https://arxiv.org/abs/2509.01135

【8】A Hybrid Ai Framework For Strategic Patent Portfolio Pruning: Integrating Learning To-Rank And Market Need Analysis For Technology Transfer Optimization
标题：战略性专利组合修剪的混合AI框架：整合按等级学习和市场需求分析以优化技术转移
链接：https://arxiv.org/abs/2509.00958

【9】ART: Adaptive Resampling-based Training for Imbalanced Classification
标题：ART：基于自适应重新采样的不平衡分类训练
链接：https://arxiv.org/abs/2509.00955

【10】Robust Spatiotemporal Forecasting Using Adaptive Deep-Unfolded Variational Mode Decomposition
标题：使用自适应深度展开变分模式分解的鲁棒时空预测
链接：https://arxiv.org/abs/2509.00703

【11】Algorithm Adaptation Bias in Recommendation System Online Experiments
标题：推荐系统在线实验中的算法自适应偏差
链接：https://arxiv.org/abs/2509.00199

【12】Adaptive Physics-Informed Neural Networks with Multi-Category Feature Engineering for Hydrogen Sorption Prediction in Clays, Shales, and Coals
标题：具有多类别特征工程的自适应物理信息神经网络用于预测粘土、页岩和煤中的氢吸收
链接：https://arxiv.org/abs/2509.00049

【13】A-FloPS: Accelerating Diffusion Sampling with Adaptive Flow Path Sampler
标题：A-FloPS：使用自适应流路采样器加速扩散采样
链接：https://arxiv.org/abs/2509.00036

【14】Transfer Learning for Minimum Operating Voltage Prediction in Advanced Technology Nodes: Leveraging Legacy Data and Silicon Odometer Sensing
标题：先进技术节点中最小工作电压预测的转移学习：利用传统数据和硅里程表传感
链接：https://arxiv.org/abs/2509.00035

强化学习(13篇)

【1】SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning
标题：SimpleTLR：用于多圈工具集成推理的端到端强化学习
链接：https://arxiv.org/abs/2509.02479

【2】Deep Reinforcement Learning for Real-Time Drone Routing in Post-Disaster Road Assessment Without Domain Knowledge
标题：没有领域知识的灾后道路评估中实时无人机路由的深度强化学习
链接：https://arxiv.org/abs/2509.01886

【3】Semi-on-Demand Transit Feeders with Shared Autonomous Vehicles and Reinforcement-Learning-Based Zonal Dispatching Control
标题：具有共享自动驾驶车辆和基于增强学习的区域调度控制的半按需交通供电器
链接：https://arxiv.org/abs/2509.01883

【4】Goal-Conditioned Reinforcement Learning for Data-Driven Maritime Navigation
标题：数据驱动海上导航的目标条件强化学习
链接：https://arxiv.org/abs/2509.01838

【5】Succeed or Learn Slowly: Sample Efficient Off-Policy Reinforcement Learning for Mobile App Control
标题：成功还是慢慢学习：针对移动应用程序控制的高效非政策强化学习示例
链接：https://arxiv.org/abs/2509.01720

【6】Reinforcement Learning for Machine Learning Engineering Agents
标题：机器学习工程代理的强化学习
链接：https://arxiv.org/abs/2509.01684

【7】The Geometry of Nonlinear Reinforcement Learning
标题：非线性强化学习的几何学
链接：https://arxiv.org/abs/2509.01432

【8】Towards High Data Efficiency in Reinforcement Learning with Verifiable Reward
标题：具有可验证奖励的强化学习中的高数据效率
链接：https://arxiv.org/abs/2509.01321

【9】Building surrogate models using trajectories of agents trained by Reinforcement Learning
标题：使用强化学习训练的代理人轨迹构建代理模型
链接：https://arxiv.org/abs/2509.01285

【10】Multi-Agent Reinforcement Learning for Task Offloading in Wireless Edge Networks
标题：无线边缘网络中用于任务卸载的多代理强化学习
链接：https://arxiv.org/abs/2509.01257

【11】Reinforcement Learning Driven Generalizable Feature Representation for Cross-User Activity Recognition
标题：用于跨用户活动识别的强化学习驱动的可推广特征表示
链接：https://arxiv.org/abs/2509.01031

【12】Reinforcement Learning of Dolly-In Filming Using a Ground-Based Robot
标题：使用地面机器人进行娃娃拍摄的强化学习
链接：https://arxiv.org/abs/2509.00564

【13】Financial Decision Making using Reinforcement Learning with Dirichlet Priors and Quantum-Inspired Genetic Optimization
标题：使用具有Dirichlet先验的强化学习和量子启发的遗传优化进行财务决策
链接：https://arxiv.org/abs/2509.00095

元学习(2篇)

【1】Learnable Loss Geometries with Mirror Descent for Scalable and Convergent Meta-Learning
标题：基于镜像下降的可学习损失几何，用于可扩展和收敛的元学习
链接：https://arxiv.org/abs/2509.02418

【2】Learning to Coordinate: Distributed Meta-Trajectory Optimization Via Differentiable ADMM-DDP
标题：学习协调：通过可区分ADMM-DDD进行分布式元轨迹优化
链接：https://arxiv.org/abs/2509.01630

符号|符号学习(1篇)

【1】Neuro-Symbolic Predictive Process Monitoring
标题：神经符号预测过程监控
链接：https://arxiv.org/abs/2509.00834

医学相关(10篇)

【1】Anisotropic Fourier Features for Positional Encoding in Medical Imaging
标题：医学成像位置编码的各向异性傅里叶特征
链接：https://arxiv.org/abs/2509.02488

【2】Baichuan-M2: Scaling Medical Capability with Large Verifier System
标题：Baichuan-M2：利用大型验证系统提升医疗能力
链接：https://arxiv.org/abs/2509.02208

【3】Content and Engagement Trends in COVID-19 YouTube Videos: Evidence from the Late Pandemic
标题：COVID-19 YouTube视频的内容和参与趋势：来自晚期大流行的证据
链接：https://arxiv.org/abs/2509.01954

【4】A Multimodal Deep Learning Framework for Early Diagnosis of Liver Cancer via Optimized BiLSTM-AM-VMD Architecture
标题：通过优化的BiLSTM-AM-VMD架构用于肝癌早期诊断的多模式深度学习框架
链接：https://arxiv.org/abs/2509.01164

【5】Multi-Modal Machine Learning Framework for Predicting Early Recurrence of Brain Tumors Using MRI and Clinical Biomarkers
标题：使用MRI和临床生物标志物预测脑肿瘤早期复发的多模式机器学习框架
链接：https://arxiv.org/abs/2509.01161

【6】Enhancing Fairness in Skin Lesion Classification for Medical Diagnosis Using Prune Learning
标题：利用Prune学习提高医学诊断皮肤病变分类的公平性
链接：https://arxiv.org/abs/2509.00745

【7】Valid Property-Enhanced Contrastive Learning for Targeted Optimization & Resampling for Novel Drug Design
标题：有效的属性增强对比学习用于新型药物设计的有针对性优化和重新排序
链接：https://arxiv.org/abs/2509.00684

【8】Automatic Screening of Parkinson's Disease from Visual Explorations
标题：通过视觉探索自动筛查帕金森病
链接：https://arxiv.org/abs/2509.01326

【9】Towards Early Detection: AI-Based Five-Year Forecasting of Breast Cancer Risk Using Digital Breast Tomosynthesis Imaging
标题：迈向早期检测：使用数字乳腺断层合成成像基于人工智能的乳腺癌风险五年预测
链接：https://arxiv.org/abs/2509.00900

【10】Can General-Purpose Omnimodels Compete with Specialists? A Case Study in Medical Image Segmentation
标题：通用全功能模型可以与专家竞争吗？医学图像分割案例研究
链接：https://arxiv.org/abs/2509.00866

蒸馏|知识提取(2篇)

【1】Distillation of a tractable model from the VQ-VAE
标题：从VQ-VAE中提炼出易于处理的模型
链接：https://arxiv.org/abs/2509.01400

【2】An Efficient GNNs-to-KANs Distillation via Self-Attention Dynamic Sampling with Potential for Consumer Electronics Edge Deployment
标题：通过自我注意力动态采样进行高效的GNNS到KAN蒸馏，具有消费电子边缘部署的潜力
链接：https://arxiv.org/abs/2509.00560

推荐(5篇)

【1】AgroSense: An Integrated Deep Learning System for Crop Recommendation via Soil Image Analysis and Nutrient Profiling
标题：AgroSense：一个通过土壤图像分析和养分分析推荐作物的集成深度学习系统
链接：https://arxiv.org/abs/2509.01344

【2】XAI-Driven Machine Learning System for Driving Style Recognition and Personalized Recommendations
标题：XAI驱动的机器学习系统，用于驾驶风格识别和个性化推荐
链接：https://arxiv.org/abs/2509.00802

【3】Game Theoretic Resilience Recommendation Framework for CyberPhysical Microgrids Using Hypergraph MetaLearning
标题：使用Hypergraph元学习的网络物理微网格博弈论弹性推荐框架
链接：https://arxiv.org/abs/2509.00528

【4】Counterfactual Risk Minimization with IPS-Weighted BPR and Self-Normalized Evaluation in Recommender Systems
标题：推荐系统中采用IPS加权BPR和自规范化评估实现反事实风险最小化
链接：https://arxiv.org/abs/2509.00333

【5】Bias Mitigation for AI-Feedback Loops in Recommender Systems: A Systematic Literature Review and Taxonomy
标题：推荐系统中人工智能反馈环的偏见缓解：系统性文献回顾和分类
链接：https://arxiv.org/abs/2509.00109

聚类(3篇)

【1】Advanced spectral clustering for heterogeneous data in credit risk monitoring systems
标题：信用风险监控系统中异构数据的高级谱聚类
链接：https://arxiv.org/abs/2509.00546

【2】Protocol for Clustering 4DSTEM Data for Phase Differentiation in Glasses
标题：用于聚集4DSTEM数据以实现眼镜中的相区分的协议
链接：https://arxiv.org/abs/2509.00943

【3】Assessing One-Dimensional Cluster Stability by Extreme-Point Trimming
标题：通过极点修剪评估一维团簇稳定性
链接：https://arxiv.org/abs/2509.00258

自动驾驶|车辆|车道检测等(3篇)

【1】Learning Social Heuristics for Human-Aware Path Planning
标题：学习社会启发法进行人性意识路径规划
链接：https://arxiv.org/abs/2509.02134

【2】Predicting NCAP Safety Ratings: An Analysis of Vehicle Characteristics and ADAS Features Using Machine Learning
标题：预测NCAP安全评级：使用机器学习分析车辆特征和ADAS特征
链接：https://arxiv.org/abs/2509.01897

【3】Robust Anomaly Detection through Multi-Modal Autoencoder Fusion for Small Vehicle Damage Detection
标题：通过多模式自动编码器融合进行稳健异常检测小型车辆损坏检测
链接：https://arxiv.org/abs/2509.01719

联邦学习|隐私保护|加密(5篇)

【1】Gaming and Cooperation in Federated Learning: What Can Happen and How to Monitor It
标题：联邦学习中的游戏与合作：会发生什么以及如何监控
链接：https://arxiv.org/abs/2509.02391

【2】Online Decentralized Federated Multi-task Learning With Trustworthiness in Cyber-Physical Systems
标题：网络物理系统中具有可信度的在线去中心化联邦多任务学习
链接：https://arxiv.org/abs/2509.00992

【3】Fairness in Federated Learning: Trends, Challenges, and Opportunities
标题：联合学习的公平性：趋势、挑战和机遇
链接：https://arxiv.org/abs/2509.00799

【4】Curriculum Guided Personalized Subgraph Federated Learning
标题：课程引导的个性化子图联邦学习
链接：https://arxiv.org/abs/2509.00402

【5】Centralized vs. Federated Learning for Educational Data Mining: A Comparative Study on Student Performance Prediction with SAEB Microdata
标题：教育数据挖掘的集中式学习与联邦学习：SAEB微数据对学生表现预测的比较研究
链接 ：https://arxiv.org/abs/2509.00086

推理|分析|理解|解释(26篇)

【1】Understanding sparse autoencoder scaling in the presence of feature manifolds
标题：了解存在特征集的情况下的稀疏自动编码器缩放
链接：https://arxiv.org/abs/2509.02565

【2】GRAM-R$^2$: Self-Training Generative Foundation Reward Models for Reward Reasoning
标题：GRAM-R$^2$：用于奖励推理的自我训练生成基础奖励模型
链接：https://arxiv.org/abs/2509.02492

【3】Understanding Space Is Rocket Science - Only Top Reasoning Models Can Solve Spatial Understanding Tasks
标题：理解空间是火箭科学-只有顶级推理模型才能解决空间理解任务
链接：https://arxiv.org/abs/2509.02175

【4】SegFormer Fine-Tuning with Dropout: Advancing Hair Artifact Removal in Skin Lesion Analysis
标题：SegFormer微调并退出：在皮肤病变分析中推进头发毛刺去除
链接：https://arxiv.org/abs/2509.02156

【5】Structure-aware Contrastive Learning for Diagram Understanding of Multimodal Models
标题：用于多峰模型图表理解的结构感知对比学习
链接：https://arxiv.org/abs/2509.01959

【6】Throttling Web Agents Using Reasoning Gates
标题：使用推理门限制Web代理
链接：https://arxiv.org/abs/2509.01619

【7】Feynman-Kac-Flow: Inference Steering of Conditional Flow Matching to an Energy-Tilted Posterior
标题：Feynman-Kac-Flow：条件流匹配到能量倾斜后验的推理引导
链接：https://arxiv.org/abs/2509.01543

【8】Evaluating the stability of model explanations in instance-dependent cost-sensitive credit scoring
标题：评估实例相关成本敏感信用评分中模型解释的稳定性
链接：https://arxiv.org/abs/2509.01409

【9】Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption
标题：具有全同形加密的实用和私有混合ML推理
链接：https://arxiv.org/abs/2509.01253

【10】An Explainable Gaussian Process Auto-encoder for Tabular Data
标题：表格数据的可解释高斯过程自动编码器
链接：https://arxiv.org/abs/2509.00884

【11】Tabular Diffusion Counterfactual Explanations
标题：表格扩散反事实解释
链接：https://arxiv.org/abs/2509.00876

【12】Exam Readiness Index (ERI): A Theoretical Framework for a Composite, Explainable Index
标题：考试准备指数（ERI）：一个综合的，可解释的指数的理论框架
链接：https://arxiv.org/abs/2509.00718

【13】Disentangling Slow and Fast Temporal Dynamics in Degradation Inference with Hierarchical Differential Models
标题：用分层差异模型解开退化推理中的慢时间动态和快时间动态
链接：https://arxiv.org/abs/2509.00639

【14】Federated Survival Analysis with Node-Level Differential Privacy: Private Kaplan-Meier Curves
标题：采用节点级差异隐私的联合生存分析：私人Kaplan-Meier曲线
链接：https://arxiv.org/abs/2509.00615

【15】Integrated Multivariate Segmentation Tree for the Analysis of Heterogeneous Credit Data in Small and Medium-Sized Enterprises
标题：集成多元细分树用于中小企业异类信用数据分析
链接：https://arxiv.org/abs/2509.00550

【16】CVPD at QIAS 2025 Shared Task: An Efficient Encoder-Based Approach for Islamic Inheritance Reasoning
标题：CVPD出席QIAS 2025共享任务：一种基于编码器的有效伊斯兰继承推理方法
链接：https://arxiv.org/abs/2509.00457

【17】SurgLLM: A Versatile Large Multimodal Model with Spatial Focus and Temporal Awareness for Surgical Video Understanding
标题：SurgLLM：具有空间焦点和时间感知的多功能大型多模式模型，用于理解手术视频
链接：https://arxiv.org/abs/2509.00357

【18】Illuminating Patterns of Divergence: DataDios SmartDiff for Large-Scale Data Difference Analysis
标题：启发分歧模式：DataDios SmartDiff用于大规模数据差异分析
链接：https://arxiv.org/abs/2509.00293

【19】Mitigating Clinician Information Overload: Generative AI for Integrated EHR and RPM Data Analysis
标题：缓解临床医生信息过载：用于集成EHR和RP数据分析的生成人工智能
链接：https://arxiv.org/abs/2509.00073

【20】From Data to Decision: A Multi-Stage Framework for Class Imbalance Mitigation in Optical Network Failure Analysis
标题：从数据到决策：光网络故障分析中缓解类失衡的多阶段框架
链接：https://arxiv.org/abs/2509.00057

【21】Probabilities of Causation and Root Cause Analysis with Quasi-Markovian Models
标题：用拟马尔科夫模型进行因果关系的可能性和根本原因分析
链接：https://arxiv.org/abs/2509.02535

【22】Using explainable artificial intelligence (XAI) as a diagnostic tool: An application for deducing hydrologic connectivity at watershed scale
标题：使用可解释人工智能（XAI）作为诊断工具：在流域尺度上推导水文连通性的应用
链接：https://arxiv.org/abs/2509.02127

【23】Inference in Spreading Processes with Neural-Network Priors
标题：具有神经网络先验的传播过程中的推理
链接：https://arxiv.org/abs/2509.02073

【24】Temporal Representation Learning for Real-Time Ultrasound Analysis
标题：实时超声分析的时间表示学习
链接：https://arxiv.org/abs/2509.01433

【25】Convergence Analysis of the PAGE Stochastic Algorithm for Convex Finite-Sum Optimization
标题：凸函数和优化的PAGE随机算法的收敛性分析
链接：https://arxiv.org/abs/2509.00737

【26】Simulation-based inference of yeast centromeres
标题：基于仿真的酵母着丝粒推断
链接：https://arxiv.org/abs/2509.00200

检测相关(11篇)

【1】ESTM: An Enhanced Dual-Branch Spectral-Temporal Mamba for Anomalous Sound Detection
标题：ESTM：用于异常声音检测的增强型双分支频谱-时间曼巴
链接：https://arxiv.org/abs/2509.02471

【2】An Efficient Intrusion Detection System for Safeguarding Radiation Detection Systems
标题：用于保障辐射检测系统的高效入侵检测系统
链接：https://arxiv.org/abs/2509.01599

【3】Securing Radiation Detection Systems with an Efficient TinyML-Based IDS for Edge Devices
标题：通过针对边缘设备的高效基于TinyML的IDS保护辐射检测系统
链接：https://arxiv.org/abs/2509.01592

【4】Detecting Rug Pulls in Decentralized Exchanges: Machine Learning Evidence from the TON Blockchain
标题：检测去中心化交易所中的地毯拉扯：来自TON区块链的机器学习证据
链接：https://arxiv.org/abs/2509.01168

【5】CCE: Confidence-Consistency Evaluation for Time Series Anomaly Detection
标题：CCE：时间序列异常检测的置信度一致性评估
链接：https://arxiv.org/abs/2509.01098

【6】NeuralSVCD for Efficient Swept Volume Collision Detection
标题：用于高效扫描体积碰撞检测的NeuralSVCD
链接：https://arxiv.org/abs/2509.00499

【7】Robust Detection of Synthetic Tabular Data under Schema Variability
标题：模式可变性下合成表格数据的鲁棒检测
链接：https://arxiv.org/abs/2509.00092

【8】Data Cartography for Detecting Memorization Hotspots and Guiding Data Interventions in Generative Models
标题：用于检测并行化热点并指导生成模型中的数据干预的数据制图
链接：https://arxiv.org/abs/2509.00083

【9】Applying Deep Learning to Anomaly Detection of Russian Satellite Activity for Indications Prior to Military Activity
标题：应用深度学习对俄罗斯卫星活动异常检测军事活动前的迹象
链接：https://arxiv.org/abs/2509.00050

【10】Automatic Pronunciation Error Detection and Correction of the Holy Quran's Learners Using Deep Learning
标题：使用深度学习自动检测和纠正《古兰经》学习者的发音错误
链接：https://arxiv.org/abs/2509.00094

【11】Exploring the Efficacy of Convolutional Neural Networks in Sleep Apnea Detection from Single Channel EEG
标题：探索卷积神经网络在单通道脑电检测睡眠呼吸暂停中的有效性
链接：https://arxiv.org/abs/2509.00012

分类|识别(10篇)

【1】L3Cube-IndicHeadline-ID: A Dataset for Headline Identification and Semantic Evaluation in Low-Resource Indian Languages
标题：L3 Cube-IndicHeadline-ID：用于低资源印度语言标题识别和语义评估的数据集
链接：https://arxiv.org/abs/2509.02503

【2】Extrapolated Markov Chain Oversampling Method for Imbalanced Text Classification
标题：不平衡文本分类的外推马尔科夫链过抽样方法
链接：https://arxiv.org/abs/2509.02332

【3】Selection of Optimal Number and Location of PMUs for CNN Based Fault Location and Identification
标题：基于CNN的故障定位和识别的最佳PFA数量和位置选择
链接：https://arxiv.org/abs/2509.02192

【4】Simulating classification models to evaluate Predict-Then-Optimize methods
标题：模拟分类模型以评估预测然后优化方法
链接：https://arxiv.org/abs/2509.02191

【5】Music Genre Classification Using Machine Learning Techniques
标题：使用机器学习技术的音乐流派分类
链接：https://arxiv.org/abs/2509.01762

【6】Causal Sensitivity Identification using Generative Learning
标题：使用生成学习进行因果敏感性识别
链接：https://arxiv.org/abs/2509.01352

【7】Speech Command Recognition Using LogNNet Reservoir Computing for Embedded Systems
标题：嵌入式系统中使用LogNNet水库计算的语音命令识别
链接：https://arxiv.org/abs/2509.00862

【8】Attribute Fusion-based Classifier on Framework of Belief Structure
标题：基于信念结构框架的属性融合分类器
链接：https://arxiv.org/abs/2509.00754

【9】Identifying Causal Direction via Dense Functional Classes
标题：通过密集功能类识别因果方向
链接：https://arxiv.org/abs/2509.00538

【10】DeepEmoNet: Building Machine Learning Models for Automatic Emotion Recognition in Human Speeches
标题：Deepspel Net：构建用于人类言语中自动情感识别的机器学习模型
链接：https://arxiv.org/abs/2509.00025

表征(4篇)

【1】A Continuous Encoding-Based Representation for Efficient Multi-Fidelity Multi-Objective Neural Architecture Search
标题：基于连续编码的高效多保真多目标神经架构搜索表示
链接：https://arxiv.org/abs/2509.01943

【2】Causal representation learning from network data
标题：从网络数据中学习因果表示
链接：https://arxiv.org/abs/2509.01916

【3】SC-GIR: Goal-oriented Semantic Communication via Invariant Representation Learning
标题：SC-GIR：通过不变表示学习实现面向目标的语义沟通
链接：https://arxiv.org/abs/2509.01119

【4】T-MLP: Tailed Multi-Layer Perceptron for Level-of-Detail Signal Representation
标题：T-MLP：用于细节级别信号表示的尾部多层感知器
链接：https://arxiv.org/abs/2509.00066

3D|3D重建等相关(2篇)

【1】LUCIE-3D: A three-dimensional climate emulator for forced responses
标题：LUCY-3D：用于强制响应的三维气候模拟器
链接：https://arxiv.org/abs/2509.02061

【2】TransForSeg: A Multitask Stereo ViT for Joint Stereo Segmentation and 3D Force Estimation in Catheterization
标题：TransForSeg：用于导管插入术中关节立体分割和3D力估计的多任务立体ViT
链接：https://arxiv.org/abs/2509.01605

编码器(3篇)

【1】Autoencoder-based non-intrusive model order reduction in continuum mechanics
标题：连续体力学中基于自动编码器的非侵入模型降阶
链接：https://arxiv.org/abs/2509.02237

【2】ReLATE: Learning Efficient Sparse Encoding for High-Performance Tensor Decomposition
标题：ReLATE：学习高效的稀疏编码以实现高性能张量分解
链接：https://arxiv.org/abs/2509.00280

【3】Quantum Circuits for Quantum Convolutions: A Quantum Convolutional Autoencoder
标题：量子卷积的量子电路：量子卷积自动编码器
链接：https://arxiv.org/abs/2509.00637

优化|敛散性(13篇)

【1】Surrogate Benchmarks for Model Merging Optimization
标题：模型合并优化的替代基准
链接：https://arxiv.org/abs/2509.02555

【2】DCPO: Dynamic Clipping Policy Optimization
标题：DCPO：动态剪裁政策优化
链接：https://arxiv.org/abs/2509.02333

【3】Threshold-Based Optimal Arm Selection in Monotonic Bandits: Regret Lower Bounds and Algorithms
标题：单调盗贼中基于阈值的最佳手臂选择：遗憾下限和算法
链接：https://arxiv.org/abs/2509.02119

【4】Differentiable Expectation-Maximisation and Applications to Gaussian Mixture Model Optimal Transport
标题：差异期望最大化及其在高斯混合模型最优运输中的应用
链接：https://arxiv.org/abs/2509.02109

【5】Privacy-Utility Trade-off in Data Publication: A Bilevel Optimization Framework with Curvature-Guided Perturbation
标题 ：数据发布中的隐私与效用权衡：具有曲线引导扰动的二层优化框架
链接：https://arxiv.org/abs/2509.02048

【6】Computational Fluid Dynamics Optimization of F1 Front Wing using Physics Informed Neural Networks
标题：利用物理信息神经网络进行F1前翼计算流体动力学优化
链接：https://arxiv.org/abs/2509.01963

【7】Globally aware optimization with resurgence
标题：具有全球意识的优化与复兴
链接：https://arxiv.org/abs/2509.01329

【8】Quantum-based QoE Optimization in Advanced Cellular Networks: Integration and Cloud Gaming Use Case
标题：高级蜂窝网络中基于量子的QOE优化：集成和云游戏用例
链接：https://arxiv.org/abs/2509.01008

【9】An Evolutionary Multi-objective Optimization for Replica-Exchange-based Physics-informed Operator Learning Network
标题：基于复制品交换的物理信息操作员学习网络的进化多目标优化
链接：https://arxiv.org/abs/2509.00663

【10】Optimized Weight Initialization on the Stiefel Manifold for Deep ReLU Neural Networks
标题：深度ReLU神经网络Stiefel Manifold上的优化权重分配
链接：https://arxiv.org/abs/2509.00362

【11】Solving Optimal Power Flow using a Variational Quantum Approach
标题：使用变分量子方法求解最优潮流
链接：https://arxiv.org/abs/2509.00341

【12】Quantum-Optimized Selective State Space Model for Efficient Time Series Prediction
标题：用于高效时间序列预测的量子优化选择性状态空间模型
链接：https://arxiv.org/abs/2509.00259

【13】Optimal information injection and transfer mechanisms for active matter reservoir computing
标题：活性物质储层计算的最佳信息注入和传输机制
链接：https://arxiv.org/abs/2509.01799

预测|估计(21篇)

【1】RDIT: Residual-based Diffusion Implicit Models for Probabilistic Time Series Forecasting
标题：RDIT：用于概率时间序列预测的基于剩余的扩散隐式模型
链接：https://arxiv.org/abs/2509.02341

【2】ST-Hyper: Learning High-Order Dependencies Across Multiple Spatial-Temporal Scales for Multivariate Time Series Forecasting
标题：ST-Hyper：跨多个时空尺度学习高层相依性以进行多元时间序列预测
链接：https://arxiv.org/abs/2509.02217

【3】Semantic and episodic memories in a predictive coding model of the neocortex
标题：新皮质预测编码模型中的语义和情景记忆
链接：https://arxiv.org/abs/2509.01987

【4】RadioDiff-Loc: Diffusion Model Enhanced Scattering Congnition for NLoS Localization with Sparse Radio Map Estimation
标题：RadioDiff-Loc：利用稀疏无线电地图估计的NLoS定位的扩散模型增强散射干扰
链接：https://arxiv.org/abs/2509.01875

【5】Optimizing In-Context Learning for Efficient Full Conformal Prediction
标题：优化上下文学习以实现高效的全共形预测
链接：https://arxiv.org/abs/2509.01840

【6】Multi-vessel Interaction-Aware Trajectory Prediction and Collision Risk Assessment
标题：多船舶相互作用感知轨迹预测和碰撞风险评估
链接：https://arxiv.org/abs/2509.01836

【7】REVELIO -- Universal Multimodal Task Load Estimation for Cross-Domain Generalization
标题：REVELIO --用于跨域概括的通用多模式任务负载估计
链接：https://arxiv.org/abs/2509.01642

【8】Entropy-Driven Curriculum for Multi-Task Training in Human Mobility Prediction
标题：人类移动预测多任务训练的信息驱动课程
链接：https://arxiv.org/abs/2509.01613

【9】From Discord to Harmony: Decomposed Consonance-based Training for Improved Audio Chord Estimation
标题：从不和谐到和谐：用于改进音频和弦估计的分解基于协和的训练
链接：https://arxiv.org/abs/2509.01588

【10】Direct Profit Estimation Using Uplift Modeling under Clustered Network Interference
标题：网络干扰下利用USYS模型估计直接利润
链接：https://arxiv.org/abs/2509.01558

【11】AT Loss: Advanced Torrential Loss Function for Precipitation Forecasting
标题：AT损失：用于降水预测的高级乌龟损失函数
链接：https://arxiv.org/abs/2509.01348

【12】StoxLSTM: A Stochastic Extended Long Short-Term Memory Network for Time Series Forecasting
标题：StoxLSTM：用于时间序列预测的随机扩展长短期记忆网络
链接：https://arxiv.org/abs/2509.01187

【13】Nonlinear Performative Prediction
标题：非线性表演预测
链接：https://arxiv.org/abs/2509.01139

【14】Crystal Structure Prediction with a Geometric Permutation-Invariant Loss Function
标题：用几何排列不变损失函数预测晶体结构
链接：https://arxiv.org/abs/2509.00832

【15】IndiaWeatherBench: A Dataset and Benchmark for Data-Driven Regional Weather Forecasting over India
标题：IndiaWeatherBench：印度数据驱动区域天气预报的数据集和基准
链接：https://arxiv.org/abs/2509.00653

【16】Progressive Element-wise Gradient Estimation for Neural Network Quantization
标题：神经网络量化的逐元素渐进梯度估计
链接：https://arxiv.org/abs/2509.00097

【17】Distribution estimation via Flow Matching with Lipschitz guarantees
标题：通过Lipschitz保证的流匹配进行分布估计
链接：https://arxiv.org/abs/2509.02337

【18】Online Complexity Estimation for Repetitive Scenario Design
标题：重复性场景设计的在线复杂性估计
链接：https://arxiv.org/abs/2509.02103

【19】An Observations-focused Assessment of Global AI Weather Prediction Models During the South Asian Monsoon
标题：南亚季风期间全球人工智能天气预测模型的以观测为中心的评估
链接：https://arxiv.org/abs/2509.01879

【20】Exploring Quantum Machine Learning for Weather Forecasting
标题：探索量子机器学习用于天气预报
链接：https://arxiv.org/abs/2509.01422

【21】MedFormer: a data-driven model for forecasting the Mediterranean Sea
标题：MedFormer：预测地中海的数据驱动模型
链接：https://arxiv.org/abs/2509.00015

其他神经网络|深度学习|模型|建模(57篇)

【1】DynaGuard: A Dynamic Guardrail Model With User-Defined Policies
标题：DynaGuard：具有用户定义策略的动态保护模型
链接：https://arxiv.org/abs/2509.02563

【2】On Transferring, Merging, and Splitting Task-Oriented Network Digital Twins
标题：面向任务的网络数字双胞胎的转移、合并和拆分
链接：https://arxiv.org/abs/2509.02551

【3】Is RL fine-tuning harder than regression? A PDE learning approach for diffusion models
标题：RL微调比回归更难吗？扩散模型的PDL学习方法
链接：https://arxiv.org/abs/2509.02528

【4】Flavors of Moonshine: Tiny Specialized ASR Models for Edge Devices
标题：Moonshine的味道：用于边缘设备的微型专用ASR模型
链接：https://arxiv.org/abs/2509.02523

【5】Fisher information flow in artificial neural networks
标题：人工神经网络中的Fisher信息流
链接：https://arxiv.org/abs/2509.02407

【6】Scaffolding Collaborative Learning in STEM: A Two-Year Evaluation of a Tool-Integrated Project-Based Methodology
标题：STEM中的协作学习框架：基于工具集成项目的方法论的两年评估
链接：https://arxiv.org/abs/2509.02355

【7】Balanced Multimodal Learning: An Unidirectional Dynamic Interaction Perspective
标题 ：平衡多模式学习：单向动态交互视角
链接：https://arxiv.org/abs/2509.02281

【8】VariAntNet: Learning Decentralized Control of Multi-Agent Systems
标题：VariAntNet：多智能体系统的学习分散控制
链接：https://arxiv.org/abs/2509.02271

【9】DaCe AD: Unifying High-Performance Automatic Differentiation for Machine Learning and Scientific Computing
标题：Dace AD：统一机器学习和科学计算的高性能自动区分
链接：https://arxiv.org/abs/2509.02197

【10】DivMerge: A divergence-based model merging method for multi-tasking
标题：DivMerge：一种用于多任务处理的基于分歧的模型合并方法
链接：https://arxiv.org/abs/2509.02108

【11】Towards Comprehensive Information-theoretic Multi-view Learning
标题：走向综合信息论的多视角学习
链接：https://arxiv.org/abs/2509.02084

【12】Genetic Programming with Model Driven Dimension Repair for Learning Interpretable Appointment Scheduling Rules
标题：具有模型驱动维度修复的遗传编程用于学习可解释的预约安排规则
链接：https://arxiv.org/abs/2509.02034

【13】BM-CL: Bias Mitigation through the lens of Continual Learning
标题：BM-CL：通过持续学习的视角缓解偏见
链接：https://arxiv.org/abs/2509.01730

【14】Constrained Decoding for Robotics Foundation Models
标题：机器人基础模型的约束解码
链接：https://arxiv.org/abs/2509.01728

【15】Distilled Pretraining: A modern lens of Data, In-Context Learning and Test-Time Scaling
标题：提炼预训练：数据、上下文学习和测试时间缩放的现代视角
链接：https://arxiv.org/abs/2509.01649

【16】Model Unmerging: Making Your Models Unmergeable for Secure Model Sharing
标题：模型分解：使您的模型无法进行安全的模型共享
链接：https://arxiv.org/abs/2509.01548

【17】Forward-Only Continual Learning
标题：仅向前推进的持续学习
链接：https://arxiv.org/abs/2509.01533

【18】CbLDM: A Diffusion Model for recovering nanostructure from pair distribution function
标题：GbLDM：从对分布函数恢复纳米结构的扩散模型
链接：https://arxiv.org/abs/2509.01370

【19】Re3: Learning to Balance Relevance & Recency for Temporal Information Retrieval
标题：Re3：学习平衡时态信息检索的相关性和近因性
链接：https://arxiv.org/abs/2509.01306

【20】Equivariant U-Shaped Neural Operators for the Cahn-Hilliard Phase-Field Model
标题：Cahn-Hilliard相场模型的等变U形神经运算符
链接：https://arxiv.org/abs/2509.01293

【21】A Class of Random-Kernel Network Models
标题：一类随机核网络模型
链接：https://arxiv.org/abs/2509.01090

【22】IMU-Enhanced EEG Motion Artifact Removal with Fine-Tuned Large Brain Models
标题：利用微调大大脑模型消除IMU增强的脑电运动预设
链接：https://arxiv.org/abs/2509.01073

【23】Chronotome: Real-Time Topic Modeling for Streaming Embedding Spaces
标题：Chronotome：流媒体嵌入空间的实时主题建模
链接：https://arxiv.org/abs/2509.01051

【24】DELTA: Variational Disentangled Learning for Privacy-Preserving Data Reprogramming
标题：Delta：用于隐私保护数据重编程的变分解纠缠学习
链接：https://arxiv.org/abs/2509.00693

【25】LLaVA-Critic-R1: Your Critic Model is Secretly a Strong Policy Model
标题：LLaVA-批评者-R1：你的批评者模型秘密地是一个强大的政策模型
链接：https://arxiv.org/abs/2509.00676