2026-08论文中文摘要
用于视觉Transformer的Grad-CAM：可解释AI中方法歧义的系统分类与审计
Grad-CAM for Vision Transformers: A Systematic Taxonomy and Audit of Methodological Ambiguity in Explainable AI
arXiv 2608.05258 · 2026-08-07正歌剧中的女高音：基于语料库的十八世纪人声类型研究
Soprano voices in opera seria: a corpus-based inquiry into eighteenth-century vocal types
arXiv 2608.05257 · 2026-08-07对抗不确定性下的态势与维持优化
Posture and Sustainment Optimization Under Adversarial Uncertainty
arXiv 2608.05256 · 2026-08-07新兴零售投资组合管理应用：带有自然语言目标的个性化、税务感知强化学习
An Emerging Retail Portfolio Management Application: Personalized, Tax-Aware Reinforcement Learning with Natural Language Goals
arXiv 2608.05255 · 2026-08-07约束优先推理：一种在数学问题求解中利用答案空间约束的无训练协议
Constraint-First Reasoning: A Training-Free Protocol for Exploiting Answer-Space Constraints in Mathematical Problem Solving
arXiv 2608.05254 · 2026-08-07超越旋转：用于表达性量化正交微调的AuroOFT
Beyond Rotations: AuroOFT for Expressive Quantized Orthogonal Fine-Tuning
arXiv 2608.05253 · 2026-08-07通过超高真空热处理降低纳米机械氮化硅谐振器的本征损耗
Reduction of intrinsic losses in nanomechanical silicon nitride resonators through thermal treatment in ultrahigh vacuum
arXiv 2608.05252 · 2026-08-07超越平衡的成核：双稳态生态系统中的入侵由前沿控制
Nucleation beyond Equilibrium: Fronts Control Invasion in Bistable Ecosystems
arXiv 2608.05251 · 2026-08-07超越全模型回滚：用于适配器状态多任务监督微调的AuroSFT
Beyond Full-Model Rollback: AuroSFT for Adapter-State Multi-Task Fine-Tuning
arXiv 2608.05250 · 2026-08-07WorldClaw：大规模智能体驱动的3D开放世界生成
WorldClaw: Agentic 3D Open-World Generation at Scale
arXiv 2608.05248 · 2026-08-07公共安全通信信号的隐藏生命周期：TETRA、TETRAPOL与P25的安全性对比分析
The Hidden Life of Public Safety Communications Signals: A Comparative Security Analysis of TETRA, TETRAPOL, and P25
arXiv 2608.05247 · 2026-08-07Search2Skill：基于 rubric 强化学习的、突破知识边界的技能蒸馏
Search2Skill: Skill Distillation Beyond Knowledge Boundaries Via Rubric-Based Reinforcement Learning
arXiv 2608.05245 · 2026-08-07用于获益-风险评估中结果排序合意性范式的统一因果推断框架
A Unified Causal Inference Framework for the Desirability of Outcome Ranking Paradigm in Benefit-Risk Evaluation
arXiv 2608.05244 · 2026-08-07边际匹配不允许因子化采样：审计因子化生成模型中的条件风格泄漏
Marginal Matching Does Not License Factorized Sampling: Auditing Conditional Style Leakage in Factorized Generative Models
arXiv 2608.05243 · 2026-08-07观测物质波中速度层级的涌现
Observing the emergence of a velocity hierarchy in matter waves
arXiv 2608.05241 · 2026-08-07一个量子比特可胜过一个比特：后训练量化的量子优势
One Qubit Can Beat One Bit: Quantum Advantage for Post-Training Quantization
arXiv 2608.05240 · 2026-08-07上下文强制：揭示自回归视频扩散中的上下文效应
In-Context Forcing: Uncovering Context Effects in Autoregressive Video Diffusion
arXiv 2608.05237 · 2026-08-07带变权区间上正交多项式的渐近行为
Asymptotics of Polynomials Orthogonal on an Interval with Varying Weights
arXiv 2608.05236 · 2026-08-07PPDL：基于大语言模型的流作为概率程序
PPDL: LLM-Based Flows as Probabilistic Programs
arXiv 2608.05234 · 2026-08-07面向连贯性的梦境场景可视化
Coherence-Oriented Dream Scene Visualisation
arXiv 2608.05233 · 2026-08-07大语言模型翻译中数值定位的分析
Analysis of Numerical Localisation in LLM Translations
arXiv 2608.05232 · 2026-08-07CyberBridge：弥合网络安全教育与行业之间的差距
CyberBridge: Bridging the Gap Between Cybersecurity Education and Industry
arXiv 2608.05231 · 2026-08-07面向可靠数据驱动时间使用优化的质量多样性方法
Quality Diversity for Reliable Data Driven Time-Use Optimization
arXiv 2608.05230 · 2026-08-07TriQua：在事实性评估中协调粒度与上下文
TriQua: Reconciling Granularity and Context in Factuality Evaluation
arXiv 2608.05228 · 2026-08-07BioMedJImpact：用于生物医学期刊AI参与度及科学影响力分析的综合数据集与大语言模型（LLM）流程
BioMedJImpact: A Comprehensive Dataset and LLM Pipeline for AI Engagement and Scientific Impact Analysis of Biomedical Journals
arXiv 2608.05227 · 2026-08-07NeuroAdaptTrainer：一款用于基于YOLO的神经元分割、交互式校正与迁移学习的Fiji/ImageJ插件
NeuroAdaptTrainer: A Fiji/ImageJ Plugin for YOLO-Based Neuron Segmentation, InteractiveCorrection and Transfer Learning
arXiv 2608.05226 · 2026-08-07Project2Task：面向自主研究的图引导项目级规划
Project2Task: Graph-Guided Project-Level Planning for Autonomous Research
arXiv 2608.05225 · 2026-08-07面向编码智能体中恶意技能文件的风险评估
Towards a Risk Assessment of Malicious Skill Files in Coding Agents
arXiv 2608.05223 · 2026-08-07Diff-Symbo：基于自回归潜在扩散模型的文本控制长时长符号音乐生成
Diff-Symbo: Text-Controlled Long-Duration Symbolic Music Generation Using Autoregressive Latent Diffusion Model
arXiv 2608.05222 · 2026-08-07轨道车辆的列车状态监测与结构健康评估系统
A System for Train Condition Monitoring and Structural Health Assessment of Rail Vehicles
arXiv 2608.05221 · 2026-08-07关于某些奇异环的动机上同调
On the motivic cohomology of some singular rings
arXiv 2608.05220 · 2026-08-07当特权指导出现错位：面向多轮智能体的状态匹配路由与语境化自蒸馏
When Privileged Guidance Misaligns: State-Matched Routing and Contextualized Self-Distillation for Multi-Turn Agents
arXiv 2608.05219 · 2026-08-07PD-GS：音素驱动的3DGS用于音频驱动的说话头生成
PD-GS: Phoneme-Driven 3DGS for Audio-Driven Talking Heads
arXiv 2608.05218 · 2026-08-07利用输入自适应优化实现视觉Transformer的对抗性效率退化研究综述
A Survey of Adversarial Efficiency Degradation for Vision Transformer by Exploiting Input-adaptive Optimization
arXiv 2608.05217 · 2026-08-07采用有限测量的MIMO-FMCW雷达进行多目标微动参数估计
Multi-Target Micro-Motion Parameter Estimation using MIMO-FMCW Radar with Limited Measurements
arXiv 2608.05216 · 2026-08-07VLAff：用于统一可操作 affordance 的视觉-语言-affordance 模型
VLAff: Vision-Language-Affordance Model for Unified Actionable Affordances
arXiv 2608.05215 · 2026-08-07马尔琴科理论、算法与应用综述
Marchenko Theory, Algorithms, and Applications: A Review
arXiv 2608.05214 · 2026-08-07StyleComposer：无需训练的多参考风格合成
StyleComposer: Training-Free Multi-Reference Style Composition
arXiv 2608.05213 · 2026-08-07SearchAuditor：长程搜索智能体失败的审计与归因
SearchAuditor: Auditing and Attributing Failures in Long-Horizon Search Agents
arXiv 2608.05212 · 2026-08-07法律援助资格与法庭结果：一种基于设计的双机器学习方法
Legal aid eligibility and court outcomes: a design-based double-machine-learning approach
arXiv 2608.05211 · 2026-08-07无辜的面板，仇恨的故事：评估与检测多轮视觉故事生成中的仇恨意图
Innocent Panels, Hateful Stories: Evaluating and Detecting Hateful Intent in Multi-Turn Visual Story Generation
arXiv 2608.05210 · 2026-08-07MapTCL：用于矢量化高清地图构建的双向对齐时序一致性学习
MapTCL: Temporal Consistency Learning via Bidirectional Alignment for Vectorized HD Map Construction
arXiv 2608.05209 · 2026-08-07三维纳维-斯托克斯湍流中极端不确定性产生事件的预测
The prediction of extreme uncertainty-production events in three-dimensional Navier-Stokes turbulence
arXiv 2608.05208 · 2026-08-07纠正特征何时有用？面向黑盒预测器的纠正特征发现智能体
When Do Corrective Features Help? An Agent for Corrective Feature Discovery on Black-Box Forecasters
arXiv 2608.05207 · 2026-08-07Otter：一种时间感知、历史条件依赖的人类国际象棋AI
Otter: A Time-Aware, History-Conditioned Human Chess AI
arXiv 2608.05206 · 2026-08-07抽象事件因果规则：归纳与应用
Abstract Event Causal Rules: Induction and Application
arXiv 2608.05205 · 2026-08-07从连续预测因子到临床阈值：基于指南的分类对缺血性卒中结局预测的性能权衡的早期证据
From Continuous Predictors to Clinical Thresholds: Early Evidence on Performance Trade-offs of Guideline-Based Categorisation for Ischaemic Stroke Outcome Prediction
arXiv 2608.05203 · 2026-08-07实践中的量子误差管理：跨栈基准测试
Quantum Error Management in Practice: A Cross-Stack Benchmark
arXiv 2608.05202 · 2026-08-07ASTELD：自主AI智能体的六轴分类框架——设计、评估与OpenClaw案例研究
ASTELD: A Six-Axis Classification Framework for Autonomous AI Agents - Design, Evaluation, and an OpenClaw Case Study
arXiv 2608.05201 · 2026-08-07等信息含量下不同基函数间引力波背景各向异性映射的等价性
Equivalence of mapping gravitational wave background anisotropy across bases at equal information content
arXiv 2608.05200 · 2026-08-07
