2026-09论文中文摘要
加权事件信号时序逻辑到无时间几何控制的可靠编译
Sound Compilation of Weighted Event Signal Temporal Logic to Timeless Geometric Control
arXiv 2609.07900 · 2026-09-29基于深度学习的扩散受限氧传感动态补偿
Computer vision enabled oxygen sensing
arXiv 2609.07625 · 2026-09-29超越矩阵符号：二次谱下降
Beyond the Matrix Sign: Quadratic Spectral Descent
arXiv 2609.07597 · 2026-09-29LRU与LFU的无遗憾混合及其最优切换成本
No-Regret Mixing of LRU and LFU with Optimal Switching Cost
arXiv 2609.07566 · 2026-09-29活性物质中稀有事件的精确哈密顿动力学
Exact Hamiltonian Dynamics of Rare Events in Active Matter
arXiv 2609.07496 · 2026-09-29关于一阶信息泛函的Brunn-Minkowski原理
On a Brunn-Minkowski principle for first-order information functionals
arXiv 2609.07463 · 2026-09-29TabBench-Bio：面向高维生物医学表格的机器学习活基准
TabBench-Bio: A Living Benchmark for Machine Learning on High-Dimensional Biomedical Tables
arXiv 2609.07441 · 2026-09-29波长均匀的量子动力学量子算法
Wavelength-Uniform Quantum Algorithms for Mixed-State Quantum Dynamics
arXiv 2609.07384 · 2026-09-29RedKnot-MLA：面向DeepSeek-V4长上下文服务的多头离线-在线复用
RedKnot-MLA: Multi-Head Offline-Online Reuse for DeepSeek-V4 Long-Context Serving
arXiv 2609.07008 · 2026-09-29面向临床医生社交媒体的医学图像识别视觉大语言基础模型
A visual large language foundational model for medical image recognition using clinician-contributed online resources
arXiv 2609.06914 · 2026-09-29Delta注意力中的精确记录遗漏：传输准则、其代价与重放证书
Exact Record Omission in Delta Attention: A Transport Criterion, Its Cost, and a Replay Certificate
arXiv 2609.06872 · 2026-09-29具有相同换位子群的自由群的子群
Subgroups of free groups with the same derived subgroup
arXiv 2609.06755 · 2026-09-29PARSER：面向长上下文LLM智能体的并行读取与深度推理
PARSER: Read in Parallel, Reason in Depth for Long-Context LLM Agents
arXiv 2609.06702 · 2026-09-29VidaForge：视频预训练数据配方的开放研究基础设施
VidaForge: Open Research Infrastructure for Video Pretraining Data Recipes
arXiv 2609.06652 · 2026-09-29GeoCo-SAVi：用于显式可编辑对象表示的几何一致性槽注意力
GeoCo-SAVi: Geometry-Consistent Slot Attention for Explicitly Editable Object Representations
arXiv 2609.06628 · 2026-09-29扩频水印模型与图像恢复的统计力学评估 II：消息解码与图像混合系统的AT稳定性
de Almeida-Thouless Stability and Replica Symmetry Breaking in a Hybrid Watermarking Model with Image and Message Decoding
arXiv 2609.06405 · 2026-09-29广义Whiting变换与次极端Kerr中波动方程的流估计
Generalized Whiting transform and Flux Estimates for Wave Equations in Subextremal Kerr
arXiv 2609.06344 · 2026-09-29两阶段随机串联系统中的1.73最优阶梯库存(R,nQ)策略
A 5/3 Guarantee for Echelon Stock (R,nQ) Policies in Two-Stage Stochastic Serial Systems
arXiv 2609.06318 · 2026-09-29关于平铺空间紧致性的自包含注记
On the topological dynamical system of tilings
arXiv 2609.06310 · 2026-09-29智能体过度信任工具：衡量对不可靠工具的依赖
Agents' Overreliance on Unreliable Tools
arXiv 2609.05587 · 2026-09-29分层条纹跟踪的实验室表征
Laboratory characterization of Hierarchical Fringe Tracking
arXiv 2609.05205 · 2026-09-29On-Policy蒸馏中关键因素是什么？数据效率与数据选择的视角
What Matters in On-Policy Distillation? A Perspective on Data Efficiency and Data Selection
arXiv 2609.05198 · 2026-09-29LIBERO-RECOVER：超越任务成功，面向机器人操纵模型的故障恢复
LIBERO-RECOVER: Beyond Task Success Towards Failure Recovery in Robotic Manipulation Models
arXiv 2609.05178 · 2026-09-29超越未压缩的压缩：RAG中软上下文压缩的两阶段训练方案
Compression Beyond the Uncompressed: A Two-Stage Training Recipe for Soft Context Compression in RAG
arXiv 2609.05152 · 2026-09-29SciDocBench：面向科学文档理解的以工作流为中心的基准与数据管道
SciDocBench: A Workflow-Centered Benchmark and Data Pipeline for Scientific Document Understanding
arXiv 2609.05141 · 2026-09-29费舍尔信息动力学：用于相空间排序的运动学框架及其在激波层和湍流中的应用
Fisher Information Dynamics: A Kinematic Framework for Phase Space Ordering with Applications to Shock Layers and Turbulence
arXiv 2609.05021 · 2026-09-29更好的理解，更好的修复？基于大语言模型的自动化程序修复中的幻觉研究
Better Understanding, Better Fixes? A Study of Hallucination in LLM-based Automated Program Repair
arXiv 2609.04909 · 2026-09-29FUSION：一款基于技能的公开可用核物理代码研究智能体
FUSION: a skill-based research agent for publicly obtainable nuclear-physics codes
arXiv 2609.04742 · 2026-09-29有限树上逆权Lin--Lu--Yau Ricci流的曲率扩散
Curvature Diffusion of Inverse-weight Lin--Lu--Yau Ricci Flow on Finite Trees
arXiv 2609.04671 · 2026-09-29用于部署后物理人工智能的持续领域自适应模型（CFAMs）
Continual Field-Adaptive Models (CFAMs) for Post-Deployment Physical AI
arXiv 2609.04552 · 2026-09-29EVOHARNESSBENCH：你的智能体能跟上不断演进的工具链吗？
EvoHarnessBench: Can Your Agents Keep Pace with an Evolving Harness?
arXiv 2609.04280 · 2026-09-29基础CNOT电路的精确综合是NP难问题
Vanilla Exact Synthesis of CNOT Circuits is NP-hard
arXiv 2609.04160 · 2026-09-29低激发射电星系与高激发射电星系的X射线吸积性质的根本差异
Fundamental differences in the X-ray accretion properties of low and high-excitation radio galaxies
arXiv 2609.04130 · 2026-09-29针对Polyak-Łojasiewicz条件的重尾一阶优化：高维极小极大界、高概率保证及固定维改进
Tight Heavy-Tailed Complexity Bounds for Optimizing Polyak--Łojasiewicz Objectives
arXiv 2609.03990 · 2026-09-29按部分奇偶性限制的超着色分拆k元组
Overcolored Partition $k$-tuples Restricted by Parity of the Parts
arXiv 2609.03926 · 2026-09-29$p$-进数上的极小近方可定义流
Minimal proximal definable flows over the $p$-adics
arXiv 2609.03873 · 2026-09-29nufftcf：基于非均匀快速傅里叶变换的非规则采样时间序列自相关与互相关函数快速估计
nufftcf: Fast Auto- and Cross-Correlation Function Estimation for Irregularly-Sampled Time Series via the Non-Uniform FFT
arXiv 2609.03866 · 2026-09-29微分同胚群对仿凯勒结构的作用：提升与在3-网诱导的动力学
Diffeomorphism Group Actions on Para-Kähler Structures: Lifts and Induced Dynamics on 3-Webs
arXiv 2609.03573 · 2026-09-29随机注意力：重新思考高效推理的KV缓存驱逐策略
Random Attention: Rethinking KV Cache Eviction for Efficient Reasoning
arXiv 2609.03430 · 2026-09-29DuplexSpeechBench-IFEval：评估全双工语音智能体的隐式指令遵循能力
DuplexSpeechBench-IFEval: Evaluating Implicit Instruction Following in Full-Duplex Voice Agents
arXiv 2609.03423 · 2026-09-29新鲜记忆，过时计划：分布式大语言模型智能体记忆的依赖范围验证
Fresh Memory, Stale Plans: Derivation Currency for Distributed LLM-Agent Memory
arXiv 2609.03340 · 2026-09-29RoboTok：用于人类演示检索与灵巧操作学习的互联网规模数据引擎
RoboTok: A Scalable Data Engine for Internet Demonstration Video Retrieval and Dexterous Manipulation Learning
arXiv 2609.03199 · 2026-09-29TRACE：用于颗粒动力学的时空接触记忆图网络模拟器
TRACE: A spatiotemporal contact memory graph network simulator for granular dynamics
arXiv 2609.02991 · 2026-09-29优化中的渗流动力学：方差级联与离散标度不变性
Percolation Dynamics in Optimization: Variance Cascades and Nested Symmetry
arXiv 2609.02373 · 2026-09-29WeaveMark：基于编码载荷扩散的鲁棒可扩展多比特大语言模型水印方案
WeaveMark: Robust and Scalable Multi-bit LLM Watermarking via Coded Payload Spreading
arXiv 2609.02177 · 2026-09-29具周期势与局域非线性的L²超临界非线性薛定谔方程的归一化解
Normalized solutions of $L^2$-supercritical NLS equations with periodic potentials
arXiv 2609.02147 · 2026-09-29通过自发对称性破缺的带毛黑洞的宇宙学意义：带毛黑洞是原初的吗？
Cosmological implications for hairy black holes via spontaneous symmetry breaking: Are Hairy Black Holes Primordial?
arXiv 2609.02102 · 2026-09-29关于保型表示的Bowditch问题与Gold猜想
From Bowditch question to Goldman conjecture for type-preserving representations
arXiv 2609.02065 · 2026-09-29对空间中平面中心构型的海森矩阵：分解、莫尔斯指数与对称性约化
The Hessian of Planar Central Configurations in Pair Space: Decomposition, Morse Index and Symmetry Reduction
arXiv 2609.01857 · 2026-09-29用于网络压缩的可复用神经基的数学理论
Linear Reusable Neural Bases Architecture for Network Compression
arXiv 2609.01550 · 2026-09-29
