2026-08论文中文摘要
利用调制器驯服树宽动态规划：图启发式算法的通用增强器
Taming Treewidth DP with Modulators: A General Booster for Graph Heuristics
arXiv 2608.04446 · 2026-08-06幂作为斐波那契和
Powers as Fibonacci Sums
arXiv 2608.04445 · 2026-08-06D²F-ReAG：面向多跳推理增强生成的动态分解与过滤
D$^2$F-ReAG: Dynamic Decomposition and Filtering for Multi-Hop Reasoning-Augmented Generation
arXiv 2608.04444 · 2026-08-06MCHA：一种面向并行-串行计算的以内存为中心的分层架构
MCHA: A Memory-Centric Hierarchical Architecture for Parallel-Sequential Computing
arXiv 2608.04443 · 2026-08-06训练动态中鲁棒性早期出现但无法保留
Robustness Emerges Early in Training Dynamics, but Is Not Preserved
arXiv 2608.04442 · 2026-08-06Season：面向基于迁移的对抗攻击的频谱感知正交梯度优化方法
Season: Spectrum-Aware Orthogonal Gradient Refinement for Transfer-Based Adversarial Attacks
arXiv 2608.04441 · 2026-08-06非齐次半线性热方程的大时间行为与成长速率
Large-time behavior and grow-up rates of inhomogeneous semilinear heat equations
arXiv 2608.04440 · 2026-08-06ExeCRE：基于执行一致性引导的自校正代码生成可靠性估计
ExeCRE: Execution-Consistency Guided Reliability Estimation for Self-Correcting Code Generation
arXiv 2608.04439 · 2026-08-06用修正恰普林气体包层遮蔽黑洞自旋：来自现象学三区时空的辐射简并
Masking Black Hole Spin with a Modified Chaplygin Gas Envelope: Radiative Degeneracies from a Phenomenological Three-Region Spacetime
arXiv 2608.04438 · 2026-08-06通过Drell-Yan产生中的角系数探测CP破坏效应
Probing the CP violation effects via the angular coefficients in Drell-Yan production
arXiv 2608.04437 · 2026-08-06ToolArtist：用于智能体图像生成的工具使用统一多模态模型
ToolArtist: Tool-Using Unified Multimodal Models for Agentic Image Generation
arXiv 2608.04436 · 2026-08-06重新审视通道有效性：基于原始视觉刺激的多维度评估
Revisiting Channel Effectiveness: A Multi-Dimensional Evaluation with Primitive Visual Stimuli
arXiv 2608.04435 · 2026-08-06OmniRouting：面向PCB布线中约束感知空间推理的语义耦合多模态基准
OmniRouting: A Semantic-Coupled Multimodal Benchmark for Constraint-Aware Spatial Reasoning in PCB Routing
arXiv 2608.04434 · 2026-08-06MERaLiON-GR：适用于英语及东南亚语言的语音性别识别模型
MERaLiON-GR: Speech Gender Recognition Model for English and SEA Languages
arXiv 2608.04433 · 2026-08-06基于局部刚度矩阵修正的低渗透屏障非拟合有限元离散裂缝模型
An unfitted finite element discrete fracture model for low-permeability barriers via local stiffness matrix modification
arXiv 2608.04431 · 2026-08-06UBLLIE：统一背光与暗光图像增强
UBLLIE: Unified Backlight and Low-Light Image Enhancement
arXiv 2608.04429 · 2026-08-06Deltoris：通过比特级稀疏性和推测性推理实现具身AI中的实时VLA推理
Deltoris: Enabling Real-time VLA Inference in Embodied AI via Bit-level Sparsity and Speculative Inference
arXiv 2608.04428 · 2026-08-06具平衡与变分不等式约束的可数广义Bregman非扩张型映射的强收敛性
Strong convergence for countable generalized Bregman nonexpansive-type mappings with equilibrium and variational inequality constraints
arXiv 2608.04427 · 2026-08-06预测再检索：从视频前缀中进行跨实例未来状态检索
Predict, Then Retrieve: Cross-Instance Future-State Retrieval from Video Prefixes
arXiv 2608.04426 · 2026-08-06SSC：用于双手操作标注的可验证结构化表示
SSC: A Verifiable Structured Representation for Bimanual Manipulation Labelling
arXiv 2608.04425 · 2026-08-06基于锚点思考：接地且高效的文档推理
Thinking with Anchors: Grounded and Efficient Document Reasoning
arXiv 2608.04424 · 2026-08-06预见不可见：叶片化石图像的非模态重建
Foreseeing the Invisible: Amodal Reconstruction of Leaf Fossil Images
arXiv 2608.04423 · 2026-08-06面向全局层级约束的激励式广告生成优化
Generative Optimization for Incentivized Advertising with Global Level Constraints
arXiv 2608.04421 · 2026-08-06通过模板固相外延在GaAs上实现基于GeSe的光学各向异性薄膜的单片集成
Monolithic integration of optically anisotropic GeSe-based films on GaAs by templated solid-phase epitaxy
arXiv 2608.04418 · 2026-08-06移位下Bruhat区间是EL-可壳化的
Shifted lower Bruhat intervals are EL-shellable
arXiv 2608.04417 · 2026-08-06主动式AI助手中个性化交互发起的偏好驱动在线适应
Preference-Driven Online Adaptation for Personalized Interaction Initiation in Proactive AI Assistants
arXiv 2608.04416 · 2026-08-06社交压力会破坏大语言模型安全评审团的多数投票机制
Social Pressure Breaks Majority Voting in LLM Safety Panels
arXiv 2608.04415 · 2026-08-06针对何种因果估计量最优？用于时变处理的纵向匹配与靶试验模拟方法的已知真值基准
Best for which estimand? A known-truth benchmark of longitudinal-matching and target-trial-emulation methods for time-varying treatments
arXiv 2608.04414 · 2026-08-06使用eBPF的微服务引导迁移的无插装依赖发现
Zero-Instrumentation Dependency Discovery for Guided Microservice Migration Using eBPF
arXiv 2608.04413 · 2026-08-06muSync-GS：面向天气与几何道路危险场景的物理同步驾驶视频合成方法
muSync-GS: Physics-Synchronized Driving Video Synthesis for Weather and Geometric Road Hazards
arXiv 2608.04412 · 2026-08-06贪心二叉搜索树具有非平凡的竞争力
The Greedy Binary Search Tree is Non-trivially Competitive
arXiv 2608.04410 · 2026-08-06量化不同的贸易质量收益
Quantifying Different Gains from Trade in Quality
arXiv 2608.04409 · 2026-08-06MESH：面向混合专家模型训练的内存高效Sinkhorn优化算法
MESH: Memory-Efficient Sinkhorn Optimization for Mixture-of-Experts Training
arXiv 2608.04407 · 2026-08-06用于异步特征子空间计算的格拉斯曼流形上的增量聚合
Incremental Aggregation on the Grassmannian for Asynchronous Eigenspace Computation
arXiv 2608.04406 · 2026-08-06基于二元主成分的无训练哈希注意力机制
Training-Free Hashing-Based Attention via Binary Principal Components
arXiv 2608.04405 · 2026-08-06Faster-WAM：面向鲁棒世界动作模型的高效推理时未来条件化方法
Faster-WAM: Efficient Inference-Time Future Conditioning for Robust World Action Models
arXiv 2608.04404 · 2026-08-06量子局部可恢复码的信息局部性
Information locality of a quantum locally recoverable code
arXiv 2608.04403 · 2026-08-06基于肘部法的MoE路由：一种用于专家选择的无需训练的推理时插件
Elbow-Based MoE Routing: A Training-Free Inference Time Plugin for Expert Selection
arXiv 2608.04401 · 2026-08-06基于牛顿-克里洛夫校正的替代模型预测实现可靠且高效的稳态计算流体动力学
Reliable and efficient steady CFD from surrogate predictions through Newton-Krylov correction
arXiv 2608.04400 · 2026-08-06未来μ子对撞机上的Bestest Little Higgs模型的顶伴侣粒子T的产生
Production of the top partner $T$ of the Bestest Little Higgs Model at the future muon collider
arXiv 2608.04399 · 2026-08-06规则手册下的近似多目标搜索
Approximate Multi-Objective Search Under Rulebooks
arXiv 2608.04398 · 2026-08-06NOLLI：用于诊断英韩性能差距的难度校准谜题基准
NOLLI: A Difficulty-Calibrated Puzzle Benchmark for Diagnosing the English-Korean Performance Gap
arXiv 2608.04397 · 2026-08-06CofactVLA：通过反事实干预消除视觉-语言-动作模型的混淆
CofactVLA: Deconfounding Vision-Language-Action Models via Counterfactual Intervention
arXiv 2608.04396 · 2026-08-06真实分数量子霍尔波函数中的零聚类几何
Zero-Clustering Geometry in Realistic Fractional Quantum Hall Wave Functions
arXiv 2608.04395 · 2026-08-06通过重新审视基于扩散模型的跨域小样本目标检测数据生成实现无额外开销的数据增强
Free-Lunch Augmentation by Revisiting Diffusion-Based Data Generation for Cross-Domain Few-Shot Object Detection
arXiv 2608.04394 · 2026-08-06当代理预测变为方程重构：因子衍生代理监督的诊断与残差学习
When Proxy Prediction Becomes Equation Reconstruction: Diagnostics and Residual Learning for Factor-Derived Proxy Supervision
arXiv 2608.04393 · 2026-08-06内接于球面的9顶点最大体积多面体
The maximum volume polytope with nine vertices inscribed in the sphere
arXiv 2608.04392 · 2026-08-06EdgeLM：面向语言模型表格理解的边缘演示
EdgeLM: Edge Demonstrations for Language Models' Table Understanding
arXiv 2608.04390 · 2026-08-06NeuroPB：利用预训练行为表示扩展神经解码
NeuroPB: Scaling Neural Decoding with Pretrained Behavioral Representations
arXiv 2608.04389 · 2026-08-06物理引导的神经艾里波束成形用于近场阻塞缓解
Physics-Guided Neural Airy Beamforming for Near-Field Blockage Mitigation
arXiv 2608.04388 · 2026-08-06
