2026-08论文中文摘要
大语言模型路由差距主要源于任务类型
Most of the LLM Routing Gap Is Task Type
arXiv 2608.23023 · 2026-08-26三维简单实Bol代数
Three-dimensional simple real Bol algebras
arXiv 2608.23015 · 2026-08-26用于物理人机交互中实时在线运动学习的自由能门控可塑性
Free-Energy-Gated Plasticity for Real-Time Online Motor Learning in Physical Human-Robot Interaction
arXiv 2608.23000 · 2026-08-26鲁比奥·德·弗朗西亚外推定理与稀疏界
Rubio de Francia's Extrapolation Theorem and Sparse Bounds
arXiv 2608.22988 · 2026-08-26SA-RSQ：一种适用于多模态推荐系统的通用稀疏表示框架
SA-RSQ: A Versatile Sparse Representation Framework for Multi-modal Recommender Systems
arXiv 2608.22979 · 2026-08-26网络干扰下因果推断中模型设定的随机化检验
Randomization tests for model specification in causal inference under network interference
arXiv 2608.22890 · 2026-08-26NemoSplat：面向介质感知水下重建的前馈式4D高斯溅射方法
NemoSplat: Feed-Forward 4D Gaussian Splatting for Media-Aware Underwater Reconstruction
arXiv 2608.22888 · 2026-08-26检索效果更好，鲁棒性更差：多跳检索增强生成（RAG）如何放大上游自动语音识别（ASR）错误
Better Retrieval, Worse Robustness: How Multi-hop RAG Amplifies Upstream ASR Errors
arXiv 2608.22872 · 2026-08-26SiZeUp：基于深度序数损失的航拍图像快速3D代理模型构建方法
SiZeUp: Fast 3D Proxy from Aerial Images via Depth Ordinal Loss
arXiv 2608.22821 · 2026-08-26概率流常微分方程中的变化检测：扩散潜在空间中的在线测试
Change Detection in Probability Flow ODE: Online Testing in Diffusion Latent Spaces
arXiv 2608.22807 · 2026-08-26OmicSync：结合证据约束大语言模型推理的可靠性感知空间多组学聚类
OmicSync: Reliability-Aware Spatial Multi-Omics Clustering with Evidence-Constrained LLM Reasoning
arXiv 2608.22785 · 2026-08-26局部图访问下的经典与量子谱密度估计
Classical and quantum spectral density estimation under local graph access
arXiv 2608.22769 · 2026-08-26受稳定子码保护的量子网络的时空路径优化
Spatio-temporal Path Optimization for Stabilizer-Code-Protected Quantum Networks
arXiv 2608.22766 · 2026-08-26d电子态密度在Pt-Ni和Pt-Pd纳米颗粒量子尺寸效应中的作用
Role of d-electron density of states in the quantum size effect of Pt-Ni and Pt-Pd nanoparticles
arXiv 2608.22743 · 2026-08-26周期性密度结构的相干性源自何处？低日冕谐振器候选体与STEREO/SECCHI的12次事件转移测试
Where Do Periodic Density Structures Acquire Coherence? A Low-Coronal Resonator Candidate and a 12-Event STEREO/SECCHI Transfer Test
arXiv 2608.22670 · 2026-08-26CausalCache：面向长视界GUI智能体的条件高保真恢复
CausalCache: Conditional High-Fidelity Restoration for Long-Horizon GUI Agents
arXiv 2608.22577 · 2026-08-26从毕达哥拉斯音乐的算术化看《几何原本》（欧几里得）第七卷中数论的诞生
The Birth of Number Theory (Book VII of Euclid's Elements) from the Arithmetization of Pythagorean Music
arXiv 2608.22572 · 2026-08-26量子计算机上复杂演化的时间反转
Time reversal of complex evolution on a quantum computer
arXiv 2608.22489 · 2026-08-26广义康托函数的局部标度与维数畸变
Local Scaling and Dimension Distortion of Generalized Cantor Functions
arXiv 2608.22423 · 2026-08-26超越普通动态近似成员的静态屏障
Beyond the Static Barrier for Ordinary Dynamic Approximate Membership
arXiv 2608.22413 · 2026-08-2687分钟完成：一种用于目标检测的无标签Vision Transformer从Softmax到线性注意力的适配方法
DiD It in 87 Minutes: A Label-Free Softmax-to-Linear Adaptation of Vision Transformers for Object Detection
arXiv 2608.22368 · 2026-08-26SANE：用于稳定极端上下文Delta规则模型的状态异常中和
SANE: State Anomaly Neutralization for Stable Extreme-Context Delta-Rule Models
arXiv 2608.22354 · 2026-08-260/1多面体顶点的近似计数：更强的困难性结果
Approximate counting of vertices of 0/1 polytopes: a stronger hardness result
arXiv 2608.22290 · 2026-08-26超越表象：揭示多模态大语言模型的情境错觉
Beyond What Meets the Eye: Unveiling Situational Illusions for Multimodal Large Language Models
arXiv 2608.22232 · 2026-08-26PURA：可证明无偏且鲁棒的多比特文本归属
PURA: Provably Unbiased and Robust Multi-Bit Watermarking for AI-Generated Text Attribution
arXiv 2608.22218 · 2026-08-26统一多模态模型中视觉生成何时能助力视觉理解？
When Does Visual Generation Help Visual Understanding in Unified Multimodal Models?
arXiv 2608.22174 · 2026-08-26图能量的秩-平均度界
Rank-Average Degree Bound for Graph Energy
arXiv 2608.22139 · 2026-08-26通过子求解器递推实现的分块稳定自适应三次正则化
Blockwise Stabilized Adaptive Cubic Regularization with Subsolvers via Recurrence
arXiv 2608.22129 · 2026-08-26FlowSep 2：用于语言查询音频源分离的自监督流匹配方法
FlowSep 2: Self-Supervised Flow Matching for Language-Queried Audio Source Separation
arXiv 2608.22111 · 2026-08-26当更多参考样本反而有害：面向少样本钢铁缺陷检测的感知污染的DINOv2记忆库
When More References Hurt: Contamination-Aware DINOv2 Memory Banks for Few-Shot Steel Defect Detection
arXiv 2608.22082 · 2026-08-26GenCoord：私有信息下的技能路径承诺
GenCoord: Skill-Path Commitments under Private Information
arXiv 2608.22055 · 2026-08-26雅可比留数公式的一个应用及关于二项式系数的若干其他猜想的证明
An application of Jacobi's residue formula and proofs of some other conjectures on binomial coefficients
arXiv 2608.22053 · 2026-08-26面向痴呆定向障碍检测与护理者感知干预的家庭社交机器人设计的利益相关者见解
Stakeholder Insights for Designing In-Home Social Robots for Dementia Disorientation Detection and Caregiver-Aware Intervention
arXiv 2608.22008 · 2026-08-26CRYSTAL程序中的Nudged Elastic Band方法：理论与应用
Nudged Elastic Band Method in the CRYSTAL code. Theory and Applications
arXiv 2608.21973 · 2026-08-26少轻子库仑系统的束缚态与共振谱
Bound and Resonant Spectra of Few-Lepton Coulomb Systems
arXiv 2608.21901 · 2026-08-26基于观测算子的像素空间扩散模型
Pixel-Space Diffusion via Observation Operators
arXiv 2608.21885 · 2026-08-26训练知识库：智能体策划文档存储的监督结构学习
Training a Knowledge Base: Supervised Structure Learning for Agent-Curated Document Stores
arXiv 2608.21829 · 2026-08-26基于t-积求解不相容张量线性系统的张量贪心双块扩展Kaczmarz方法
A Tensor Greedy Double-Block Extended Kaczmarz Method for Inconsistent Tensor Linear Systems under the t-product
arXiv 2608.21705 · 2026-08-26再探静电持久长度 I. 理论
Electrostatic Persistence Length Revisited. I. Theory
arXiv 2608.21617 · 2026-08-26利用穿越宇宙缪子重建表面液态氩时间投影室的高保真光产额图
Reconstructing High-Fidelity Light Yield Maps for Surface LArTPCs Using Crossing Cosmic Muons
arXiv 2608.21581 · 2026-08-26可自组装闪锌矿光子晶体中的外尔点与费米弧表面态
Weyl Points and Fermi Arc Surface States in a Self-assemblable Zinc-Blende Photonic Crystal
arXiv 2608.21565 · 2026-08-26NestyNet. IV. 无预先选定的定律
NestyNet. IV. Laws Chosen by Nothing in Advance
arXiv 2608.21491 · 2026-08-26用于高斯混合熵的高斯-埃尔米特求积法及动作空间埃尔米特代理模型
Gauss--Hermite Quadrature for Gaussian-Mixture Entropy with an Action-Space Hermite Surrogate
arXiv 2608.21467 · 2026-08-26Tavis-Cummings模型中的集体光子回波：分布无关性、两种失谐机制与Dicke梯
Collective photon echoes in the Tavis-Cummings model: distribution independence, two detuning regimes, and the Dicke ladder
arXiv 2608.21442 · 2026-08-26基于卡尔曼滤波的信息融合中惯性测量单元（IMU）参数的设置
The Setting of IMU Parameters in Kalman Filtering-based Information Fusion
arXiv 2608.21433 · 2026-08-26用于评估大型语言模型（LLMs）有限理性的k级可区分机制
Level-k Distinguishable Mechanisms for Evaluating Bounded Rationality in LLMs
arXiv 2608.21296 · 2026-08-26视觉提示就足够了？研究渐进式视觉支架下视觉语言模型的空间推理能力
Is Visual Prompting All You Need? Studying VLM Spatial Reasoning under Progressive Visual Scaffolds
arXiv 2608.21170 · 2026-08-26NestyNet. III. 基于解析神经代理的符号回归
NestyNet. III. Symbolic Regression from Analytic Neural Surrogates
arXiv 2608.21051 · 2026-08-26识别逻辑映射中动力学转变的结构
Identifying the structure of dynamical transitions in logistic map
arXiv 2608.20956 · 2026-08-26去噪未来：用于时序知识图谱外推的上下文感知谱扩散
Denoising the Future: Context-Aware Spectral Diffusion for Temporal Knowledge Graph Extrapolation
arXiv 2608.20804 · 2026-08-26
