2026-09论文中文摘要
测试具有多项式查询复杂度的二元秩
Testing the Binary Rank with Polynomial Query Complexity
arXiv 2609.10496 · 2026-09-30关于 Johnson 和 Kollár 的两个开放问题
On Two Open Problems by Johnson and Kollár
arXiv 2609.10481 · 2026-09-30开Calabi-Yau流形的形变族与Stein性质
Deformation families of open Calabi-Yau manifolds and Steinness
arXiv 2609.10454 · 2026-09-30关于中间quandle着色链环不变量与因果性检测
On Medial Quandle Coloring Link Invariant and Detecting Causality
arXiv 2609.10449 · 2026-09-30无教师自蒸馏一致性轨迹学习用于快速语音增强
Teacher-Free Self-Distilled Consistency Trajectory Learning for Fast Speech Enhancement
arXiv 2609.10392 · 2026-09-30量化超均匀和非超均匀多粒子系统中的平移与键取向有序度量
Quantifying translational and bond-orientational order metrics in hyperuniform and nonhyperuniform many-particle systems
arXiv 2609.10359 · 2026-09-30通过Enriques Cremona变换的双有理三次四fold
Birational cubic fourfolds via an Enriques Cremona transformation
arXiv 2609.10353 · 2026-09-30量子信道编码的具有二次收敛性的平方和层级
A Sum-of-Squares Hierarchy with Quadratic Convergence for Quantum Channel Coding
arXiv 2609.09629 · 2026-09-30AgentGrad：面向多智能体系统的干预引导提示优化
AgentGrad: Intervention-guided Prompt Optimization for Multi Agent Systems
arXiv 2609.08572 · 2026-09-30TASG-Explore：面向不平坦地形的地面机器人可通行性感知扇区引导探索
TASG-Explore: Traversability-Aware Sector-Guided Exploration for Ground Robot on Uneven Terrain
arXiv 2609.08512 · 2026-09-30SRPO：用于多智能体大语言模型的集合级相对策略优化
SRPO: Setwise Relative Policy Optimization for Multi-Agent Systems
arXiv 2609.08452 · 2026-09-30TV正则化的在线策略蒸馏：方向在在线策略蒸馏中至关重要
TV-Regulated OPD: Direction Matters in On-Policy Distillation
arXiv 2609.08341 · 2026-09-30Kruglov弦的Bogomol'nyi方程
Bogomol'nyi equations for Kruglov strings
arXiv 2609.08111 · 2026-09-30卡尔曼Delta网络：不确定性感知的联想记忆
Kalman Delta Networks: Uncertainty-aware Associative Memory
arXiv 2609.07816 · 2026-09-30X-DigCheck：共同演化应用画像与知识图谱，以Rupe Magna的RTI文档为例
X-DigCheck: Co-Evolving Application Profiles and Knowledge Graphs, Demonstrated on the RTI Documentation of Rupe Magna
arXiv 2609.07694 · 2026-09-30在线代理修复：在闭环发现中将高保真反馈与搜索长度解耦
Online Surrogate Repair: Decoupling High-Fidelity Feedback from Search Length in Closed-Loop Discovery
arXiv 2609.07655 · 2026-09-30沿矩曲线的多线性希尔伯特变换的有界性
Boundedness of Multilinear Hilbert Transforms Along Moment Curves
arXiv 2609.07513 · 2026-09-30平滑Picard哈密顿蒙特卡洛
Smoothed Picard Hamiltonian Monte Carlo
arXiv 2609.06906 · 2026-09-30Hadamard空间中具有非紧闭凸包的有限集
A finite set with noncompact closed convex hull in a Hadamard space
arXiv 2609.06455 · 2026-09-30CST-WM：用于具身视觉跟踪的因果结构世界模型
CST-WM: A Causally Structured World Model for Embodied Visual Tracking
arXiv 2609.06302 · 2026-09-30通过重启朗之万实现快速PAC全局优化：探索、利用与退化冷却
Fast PAC Global Optimization via Restarted Langevin: Exploration, Exploitation, and Degenerate Cooling
arXiv 2609.06196 · 2026-09-30超越免重训练的MoE压缩：压缩后调整的成本归一化研究
Beyond Retraining-Free MoE Compression: A Cost-Normalized Study of Post-Compression Adjustment
arXiv 2609.06076 · 2026-09-30具有均匀随机步长的Metropolis调整Langevin算法的全局谱间隙
A spectral gap for Metropolis-adjusted Langevin algorithm with a uniformly randomized step size
arXiv 2609.05847 · 2026-09-30湍流的规范理论
A Gauge Theory of Turbulence:
arXiv 2609.05544 · 2026-09-30SimFuse3D：面向跨平台3D目标检测的源引导目标模拟与置信度引导多阶段定位重加权
SimFuse3D: Source-Guided Target Simulation and Confidence-Guided Multi-Stage Localization Reweighting for Cross-Platform 3D Object Detection
arXiv 2609.04886 · 2026-09-30用于具身智能体高效长时程空间记忆的语言轨迹编码
Linguistic Trajectory Encoding for Efficient Long-Horizon Spatial Memory in Embodied Agents
arXiv 2609.04802 · 2026-09-30最后翻译基准
Last Translation Benchmark
arXiv 2609.04173 · 2026-09-30基于训练子样本的模型辅助估计：一种结合设计方差估计的两阶段抽样方法
Model-assisted estimation with a training subsample: a two-phase sampling approach with design-based variance estimation
arXiv 2609.04082 · 2026-09-30分裂酉Shimura簇的Oort猜想
Oort's conjecture for split unitary Shimura varieties
arXiv 2609.04072 · 2026-09-30离线强化学习中的多步近端策略改进
Is One Step Enough for Offline Policy Improvement?
arXiv 2609.03842 · 2026-09-30边际而非窗口：无训练的逐步有损推测解码
Margins, Not Windows: Training-Free Per-Step Lossy Speculative Decoding
arXiv 2609.02897 · 2026-09-30类谱无网格离散化学习
Learning Spectrally Optimised Mesh-Free Discretisations
arXiv 2609.02833 · 2026-09-30MeshSplatBench：基于三角形的神经渲染统一基准
MeshSplatBench: A Unified Benchmark for Triangle- and Mesh-Based Neural Rendering
arXiv 2609.01306 · 2026-09-30针对多重反向互补与回文重复的编码
Coding for Multiple Reverse-Complement and Palindromic Duplications
arXiv 2609.00779 · 2026-09-30适可而止：机器翻译重排序中高效候选生成的弃权策略
Quit While You're Ahead: Quit for Efficient Candidate Generation in Machine Translation Reranking
arXiv 2609.00588 · 2026-09-30谱Barron空间中带状态依赖扩散的有限时间域Hamilton-Jacobi-Bellman方程
Variable-Coefficient Parabolic Equations and Finite-Horizon Hamilton--Jacobi--Bellman Equations in Augmented Spectral Barron Spaces
arXiv 2609.00529 · 2026-09-30quandle着色与内自同构的R等价
R-equivalence of quandle colorings and inner automorphisms
arXiv 2608.30132 · 2026-09-30多类队列调度中的审查周期敏感性
Review-Period Sensitivity in Multiclass Queue Scheduling
arXiv 2608.29398 · 2026-09-30多轮人机对话中的善意偏见
Benevolent Bias in Multi-Turn Human-Agent Dialogue
arXiv 2608.29206 · 2026-09-30Flow-JEPA：用于JEPA世界模型中鲁棒潜在动力学的流匹配方法
Flow-JEPA: Robust Latent Dynamics for JEPA World Models via Flow Matching
arXiv 2608.29029 · 2026-09-30Hydra：一种具有离散潜在规划与连续流匹配执行能力的导航世界动作模型
Hydra: A Navigation World Action Model with Discrete Latent Planning and Continuous Flow-Matching Execution
arXiv 2608.28995 · 2026-09-30替代的错觉：在基础模型时代重新审视专用机器学习模型
The Illusion of Replacement: Rethinking Specialized Machine Learning Models in the Foundation Model Era
arXiv 2608.28980 · 2026-09-30验证工具的覆盖范围决定其价值：AI编码智能体中验证面、产物质量与成本的对照研究
The reach of a verification tool decides its value: A controlled study of verification surface, artifact quality, and cost in AI coding agents
arXiv 2608.28795 · 2026-09-30来自AdS$_3$上广义相对论的全息扭曲共形场论（CFTs）
Holographic warped CFTs from General Relativity on AdS$_3$
arXiv 2608.28751 · 2026-09-30超出刚性与有限性条件下，由顶点算子代数构造带奇点的模函子
Modular Functors with Singularities from Vertex Operator Algebras Beyond Rigidity and Finiteness
arXiv 2608.28579 · 2026-09-30CultureConverse：面向东亚与东南亚文化适配助手的多语言多轮模拟工具包
CultureConverse: A Multilingual Multi-turn Simulation Harness for Culturally Grounded Assistance in East and Southeast Asia
arXiv 2608.28405 · 2026-09-30表演式隐私：当差分隐私最大化效用时
Performative Privacy: When Differential Privacy Maximizes Utility
arXiv 2608.28198 · 2026-09-30流形上熵的凹性及其他性质
Concavity and other properties of the entropy on manifolds
arXiv 2608.27327 · 2026-09-30ClusterAttention：一种无需训练的双向注意力加速方法
ClusterAttention: A training-free speedup of bidirectional attention
arXiv 2608.26965 · 2026-09-30用于等变图生成的格罗莫夫-蒙日流匹配
Alignment Matters Inside and Out in Equivariant Graph Flow Matching
arXiv 2608.26961 · 2026-09-30
