2026-07论文中文摘要
音乐源分离训练（MSST）：训练和评估音乐分离模型的统一框架
Music-Source-Separation-Training (MSST): A Unified Framework for Training and Evaluating Music Demixing Models
arXiv 2607.23395 · 2026-07-28用于减轻大语言模型微调中隐藏行为的推理时共识
Inference-Time Consensus for Mitigating Hidden Behaviors from LLM Fine-Tuning
arXiv 2607.23394 · 2026-07-28关键区间A*：通过结构抽象加速网格路径查找
Key-Interval A*: Accelerating Grid Pathfinding via Structural Abstraction
arXiv 2607.23393 · 2026-07-28超行列式波函数
Hyperdeterminant wavefunctions
arXiv 2607.23392 · 2026-07-28在用于格心离散化的多重网格公理下，对称V循环可能发散
The symmetric V-cycle can diverge under the multigrid axioms for cell-centred discretisations
arXiv 2607.23391 · 2026-07-28深度何时能取代精度？量化神经计算的资源理论
When Can Depth Replace Precision? A Resource Theory of Quantized Neural Computation
arXiv 2607.23390 · 2026-07-28在真实芯片上渲染：GPU渲染时间作为一种被动的、抗人工智能的验证码信号
Rendering on Real Silicon: GPU Render-Timing as a Passive, AI-Resistant CAPTCHA Signal
arXiv 2607.23389 · 2026-07-28关于史瓦西 - 德西特时空中引力排斥对测地线偏差的影响
On the effect of gravitational repulsion on geodesic deviation in the Schwarzschild-de Sitter spacetime
arXiv 2607.23387 · 2026-07-28确信错误：前沿大语言模型规则评估中的异常链崩溃
Confidently Wrong: Exception Chain Collapse in Frontier LLM Rule Evaluation
arXiv 2607.23386 · 2026-07-28关于均匀秩三定向拟阵中的平衡回路
On balanced circuits in uniform rank-three oriented matroids
arXiv 2607.23385 · 2026-07-28语义半增量无数据关联目标 SLAM
Semantic Semi-Incremental Data-Association-Free Object SLAM
arXiv 2607.23384 · 2026-07-28生物分子凝聚物中的稳态噪声缓冲取决于由界面和液滴大小效应调制的相多重性
Homeostatic Noise Buffering in Biomolecular Condensates Hinges on Phase Multiplicity Modulated by Interfacial and Droplet Size Effects
arXiv 2607.23382 · 2026-07-28Bestvina-Brady群的自同构：IA刚性、算术共轭性与有限性
Automorphisms of Bestvina-Brady Groups: IA Rigidity, Arithmetic Commensurability, and Finiteness
arXiv 2607.23380 · 2026-07-28当激活预言机学会不读取时：微调预言机中的特定概念盲点
When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles
arXiv 2607.23379 · 2026-07-28李代数同态的上同调维数
Cohomological dimension of a Lie algebra homomorphism
arXiv 2607.23378 · 2026-07-28用于无振动光机械传感的双质量毫克级扭转振荡器
Dual mass milligram-scale torsion oscillator for vibration-free optomechanical sensing
arXiv 2607.23376 · 2026-07-28一个可行宇宙所需的微调程度：不脆弱吗？
The Degree of Fine-Tuning Needed for a Viable Universe: Not Fragile?
arXiv 2607.23374 · 2026-07-28UltraViT：用于大型视觉语言模型的延迟优化设备端视觉编码器
UltraViT: Latency-Optimized On-device Vision Encoder for Large Vision-Language Models
arXiv 2607.23373 · 2026-07-28研究用于血管分割的卷积神经网络的视觉线索：显微镜和眼底成像的案例研究
Investigating the Visual Cues of CNNs for Vascular Segmentation: A Case Study in Microscopy and Fundus Imaging
arXiv 2607.23371 · 2026-07-28通过社会情绪与技术特征的状态感知多模态融合预测比特币价格走势
Bitcoin Price Direction Prediction via Regime-Aware Multi-Modal Fusion of Social Sentiment and Technical Features
arXiv 2607.23370 · 2026-07-28用基于树图解析支持的加权班扎夫交互解释生物医学CLIP
Explaining BiomedCLIP with Weighted Banzhaf Interactions Supported by Tree-Gram Parsing
arXiv 2607.23368 · 2026-07-28具有严格递增估值的公平分配：双主体 EF1 和帕累托最优的精确阈值
Fair Division with Strictly Increasing Valuations: A Tight Threshold for Two-Agent EF1 and PO
arXiv 2607.23367 · 2026-07-28码本容量决定分层离散视频压缩中不同分辨率下的感知质量
Codebook Capacity Governs Perceptual Quality Across Resolutions in Hierarchical Discrete Video Compression
arXiv 2607.23366 · 2026-07-28关于人工智能系统工程中的人工智能安全与保障技术债务
On AI Safety and Security Technical Debt in Engineering AI-Enabled Systems
arXiv 2607.23365 · 2026-07-28关于使用结果奖励进行无偏且长度不变的策略优化的不可能性
On the Impossibility of Unbiased and Length-Invariant Policy Optimization with Outcome Rewards
arXiv 2607.23364 · 2026-07-28金属-电介质超材料中的二阶非线性能否为难以捉摸的光子时间晶体铺平道路？
Can Second-Order Nonlinearity in Metal-Dielectric Metamaterials Pave a Way Toward Elusive Photonic Time Crystals?
arXiv 2607.23363 · 2026-07-28贪心最长匹配分词的联合优化
Joint Optimization for Greedy Longest-match Tokenization
arXiv 2607.23362 · 2026-07-28语言生成中的幻觉率
Hallucination Rates in Language Generation
arXiv 2607.23361 · 2026-07-28全局博弈中的低秩收益与极限唯一性
Low-Rank Payoffs and Limit Uniqueness in Global Games
arXiv 2607.23360 · 2026-07-28定价潘多拉魔盒：顺序信息获取中的收益最大化
Pricing Pandora's Boxes: Revenue Maximization in Sequential Information Acquisition
arXiv 2607.23359 · 2026-07-28Fanpy 2.0：用于灵活变分设计的波函数实现与分析工具
Fanpy 2.0: Wavefunction Implementation and Analysis Tools for Flexible Ansatz Design
arXiv 2607.23358 · 2026-07-28有向无环图的量级同调与欧拉特征
Magnitude homology and Euler characteristics of directed acyclic graphs
arXiv 2607.23357 · 2026-07-28如何绘制平面图：一项实验评估
How to Draw a Planar Graph: An Experimental Evaluation
arXiv 2607.23356 · 2026-07-28通过结合Transformer和程序依赖图增强代码理解以进行影响分析
Enhancing Code Understanding for Impact Analysis by Combining Transformers and Program Dependence Graphs
arXiv 2607.23355 · 2026-07-28可扩展地计算度量量级
Scalably computing metric magnitude
arXiv 2607.23354 · 2026-07-28单项式序列的朱利亚集
Julia Sets for Sequences of Monomials
arXiv 2607.23353 · 2026-07-28不可压缩曲面、层次结构与平凡纽结识别
Incompressible surfaces, hierarchies and unknot recognition
arXiv 2607.23350 · 2026-07-28对应用于多数规则下社会系统的玻尔兹曼机生成能力的探索
Exploration of the generative capabilities of Boltzmann machines applied to social systems under the majority rule
arXiv 2607.23349 · 2026-07-28用于混合连续-分类表格数据的Logit坐标生成模型
Logit-Coordinate Generative Models for Mixed Continuous-Categorical Tabular Data
arXiv 2607.23348 · 2026-07-28用于收缩图像重建的可训练非扩张去噪器
Trainable Nonexpansive Denoisers for Contractive Image Reconstruction
arXiv 2607.23347 · 2026-07-28SPRKD：通过鞍点近似实现深度神经网络的有效知识蒸馏
SPRKD: Effective Knowledge Distillation for Deep Neural Networks via Saddle Region Approximation
arXiv 2607.23346 · 2026-07-28具有耦合积的置换上的霍普夫代数
A Hopf algebra on permutations with a coupling product
arXiv 2607.23345 · 2026-07-28基于BERT的模型与大语言模型在低资源命名实体识别中的比较研究：以马拉地语为例
BERT-based Models vs. Large Language Models for Low-Resource Named Entity Recognition: A Comparative Study on Marathi
arXiv 2607.23344 · 2026-07-28用于高效患者特异性术中二维/三维配准的患者无关合成预训练
Patient-Agnostic Synthetic Pretraining for Efficient Patient-Specific Intraoperative 2D/3D Registration
arXiv 2607.23343 · 2026-07-28探索将OODA循环作为应对冲突的系统思维方式
Exploring the OODA Loop as a Systematic Way of Thinking in Coping with Conflicts
arXiv 2607.23342 · 2026-07-28ABISS：通过代理交互评估文本到SQL系统
ABISS: Evaluating Text-to-SQL Systems Through Agent Interaction
arXiv 2607.23340 · 2026-07-28解释堆叠动力学SZ效应I：速度重建和非线性速度效应
Interpreting the stacked kinetic SZ effect I: velocity reconstruction and non-linear velocity effects
arXiv 2607.23339 · 2026-07-28图压缩是否保留信号传播？
Does Graph Compression Preserve Signal Propagation?
arXiv 2607.23338 · 2026-07-28从异构轨迹中发现神经算子
Neural operator discovery from heterogeneous trajectories
arXiv 2607.23337 · 2026-07-28建筑环境中人工智能代理社会的宪法治理：一项研究议程
Constitutional governance for societies of AI agents in the built environment: a research agenda
arXiv 2607.23336 · 2026-07-28
