2026-09论文中文摘要
生成式图像模型中的持久身份保留：基准与评估系统
Persistent Identity Preservation in Generative Image Models: A Benchmark and Evaluation System
arXiv 2609.04151 · 2026-09-11Barnacle：面向基于DAG的共识的自适应多领导者调度
Barnacle: Adaptive Multi-Leader Scheduling for DAG-Based Consensus
arXiv 2609.03978 · 2026-09-11图平衡问题的Wang-Sitters舍入方案的11/6上确界
The 11/6 supremum of the Wang-Sitters rounding scheme for graph balancing
arXiv 2609.03890 · 2026-09-11一类具有间接信号产生的临界拟线性趋化模型有界性的尖锐质量阈值：径向对称情形
A Sharp Mass Threshold for Boundedness in a Critical Quasilinear Chemotaxis Model With Indirect Signal Production: The Radially Symmetric Case
arXiv 2609.03713 · 2026-09-11中心扭曲的格里博夫谱与精细格里博夫-兹万齐格框架中的有限体积高斯响应
Center-twisted Gribov spectra and the finite-volume Gaussian response in the refined Gribov--Zwanziger framework
arXiv 2609.03525 · 2026-09-11RecurTrace：基于循环时间记忆的自适应潜在推理
RecurTrace: Adaptive Latent Reasoning with Loop-Time Memory
arXiv 2609.03379 · 2026-09-11面向参与式民主的偏好推断的集体中心评估
Toward Collective-Centric Evaluation of Preference Inference for Participatory Democracy
arXiv 2609.02990 · 2026-09-11VibeVoice-ASR-Streaming 技术报告
VibeVoice-ASR-Streaming Technical Report
arXiv 2609.02812 · 2026-09-11自旋轨道纠缠诱导的磁八极拉什巴-埃德尔斯坦效应的涌现
Emergence of the magnetic octupole Rashba-Edelstein effect from spin-orbit entanglement
arXiv 2609.02768 · 2026-09-11H₂分子光电离的计算方法：一项比较研究
Accurate theoretical methods for photoionization of H$_2$ molecules
arXiv 2609.02461 · 2026-09-11标量活性物质中守恒伊辛临界性的非平衡修正：Ward恒等式、能谱与长交叉行为
Nonequilibrium corrections to conserved Ising criticality in scalar active matter: Ward identities, spectrum, and long crossovers
arXiv 2609.02351 · 2026-09-11OmegaUse-SOP：基于人类演示的专业计算机使用SOP工程
OmegaUse-SOP: SOP Engineering for Professional Computer Use from Human Demonstrations
arXiv 2609.02149 · 2026-09-11具有数量依赖型奖励的多战场竞赛中的均衡结构
Equilibrium Architecture in Multi-Battle Contests with Count-Dependent Prizes
arXiv 2609.02031 · 2026-09-11OUTLETS：基于推测解码主干的输出长度预测
OUTLETS: Output-Length Prediction from Speculative Decoding Backbones
arXiv 2609.01068 · 2026-09-11SQUID端接超导腔中的高频引力波换能
High-Frequency Gravitational-Wave Transduction in a SQUID-Terminated Superconducting Cavity
arXiv 2609.01054 · 2026-09-11Streaming4D：通过分块视频生成与增量重建加速4D世界模型
Streaming4D: Accelerate 4D World Models via Block-wise Video Generation and Incremental Reconstruction
arXiv 2609.00610 · 2026-09-11学习保留什么：用于多智能体大语言模型系统高效协作的门控记忆路由
Learning What to Retain: Gated-Memory Routing for Efficient Collaboration in Multi-Agent LLM Systems
arXiv 2609.00237 · 2026-09-11探测小x区域的致密核物质：一种全局分析框架的工作流程
Probing Dense Nuclear Matter at Small-x: A workflow for a global analysis framework
arXiv 2609.00230 · 2026-09-11金星基准：利用DAVINCI解决类地系外行星光谱中的简并性问题
The Venus Benchmark: Resolving Degeneracies in Terrestrial Exoplanet Spectra with DAVINCI
arXiv 2609.00147 · 2026-09-11REAL-Q：基于动态梯度下降的端到端大语言模型量化方法
REAL-Q: E2E LLM Quantization via Dynamic Gradient Descent
arXiv 2609.00049 · 2026-09-11约束系统中可激发零模的高效搜索
Efficient search for excitable zero-modes in constrained systems
arXiv 2608.31165 · 2026-09-11带角速度的大N矢量模型
The large $N$ vector model with angular velocity
arXiv 2608.31151 · 2026-09-11关于 queer 超 Yangian ${\rm Y}(\frak{q}_1)$ 的中心的注记
A note on the center of the queer super Yangian ${\rm Y}(\mathfrak{q}_1)$
arXiv 2608.30293 · 2026-09-11Motus2：一种用于灵巧操作的自进化通用世界模型
Motus2: A Self-Evolving General World Model for Dexterous Manipulation
arXiv 2608.30237 · 2026-09-11循环归纳中的除数格与舒尔正性
The cyclic-induction Schur cone: Boolean sums, Ramanujan-square positivity, and integral structure
arXiv 2608.29757 · 2026-09-11前馈三维重建的置信度校准审计
A Calibration Audit of Confidence in Feed-Forward 3D Reconstruction Models
arXiv 2608.29705 · 2026-09-11ICEGR：面向电商搜索的意图一致型端到端生成式检索框架
ICEGR: An Intent-Coherent End-to-End Generative Retrieval Framework for E-commerce Search
arXiv 2608.29652 · 2026-09-11商空间上的量子自然梯度
Quantum Natural Gradient on Quotient Spaces
arXiv 2608.29350 · 2026-09-11广义扬岑过滤、双曲限制与特征闭链
Generalized Jantzen filtration, hyperbolic restriction, and characteristic cycles
arXiv 2608.29312 · 2026-09-11面向汉藏语系语言的跨语言罗马字生态系统：普通话-粤语配对案例研究
Toward a Cross-Lingual Romanization Ecosystem for Sinitic Languages: A Paired Mandarin-Cantonese Case Study
arXiv 2608.29170 · 2026-09-11恰当评分规则如何塑造大语言模型（LLM）的预测
How Proper Scoring Rules Shape LLM Forecasting
arXiv 2608.28482 · 2026-09-11ITER：面向智能体搜索的交互感知检索
ITER: Interaction-Aware Retrieval for Agentic Search
arXiv 2608.27912 · 2026-09-11关于数值半径范数下正交性的对称性
On symmetricity of orthogonality with respect to numerical radius norm
arXiv 2608.27608 · 2026-09-11调和方法中的4D N=2超对称的卡西米尔算子
Casimir operators of $4D\,, \mathcal{N}=2$ supersymmetry in the harmonic approach
arXiv 2608.27412 · 2026-09-11具二次多变量的函数域上的Birch定理
Birch's theorem over function fields with quadratically many variables
arXiv 2608.27285 · 2026-09-11基于深度学习图像质量度量的相衬显微CT可微分抖动校正方法
Differentiable Jitter Correction using Deep Learning-based Image Quality Metric for Phase-Contrast Micro-CT
arXiv 2608.27034 · 2026-09-11面向低空无线网络的频谱与空域资源联合优化
Joint Spectrum and Airspace Resource Optimization for Low-altitude Wireless Network
arXiv 2608.26830 · 2026-09-11FaultLens：为生成的操作程序学习紧凑的行为测试套件
FaultLens: Learning Compact Behavioral Test Suites for Generated Operational Programs
arXiv 2608.26746 · 2026-09-11GameWAM：面向电子游戏的世界动作模型
GameWAM: A World Action Model for Video Games
arXiv 2608.26200 · 2026-09-11通过稳定对构造拓扑弦爆破方程
Topological String Blowup Equations via Stable Pairs
arXiv 2608.25397 · 2026-09-11CAT-GS：通过校准门控与融合操作实现平衡多模态学习
CAT-GS: Balanced Multimodal Learning via Calibrated Gating and Fusion Surgery
arXiv 2608.24947 · 2026-09-11有限域上等变椭圆同态对应的有理函数的置换性与例外性
Indecomposable rational functions over finite fields with Galois closure of genus one: Reconstruction and exceptionality
arXiv 2608.24744 · 2026-09-11过程监控器预测的因果解释
Causal Explanations of Process Monitor Predictions
arXiv 2608.24672 · 2026-09-11随机加权均值的幂和收敛性与Breiman猜想
Breiman's conjecture and normalized jumps of subordinators
arXiv 2608.24616 · 2026-09-11可交换对Hawkes参考模型与硬球碰撞统计的累积量障碍
An Exchangeable Pair-Hawkes Reference Model and a Cumulant Obstruction for Hard-Sphere Collision Statistics
arXiv 2608.24595 · 2026-09-11在逆序列上迭代莱默码：卡塔兰不动点与有限稳定化
Iterating the Lehmer code on inversion sequences: Catalan fixed points and finite stabilization
arXiv 2608.24476 · 2026-09-11单义性的回归：专用神经元提升多层感知机的数据效率
Revenge of Monosemanticity: Neuron Specialization as a New Form of Feature Learning in MLPs
arXiv 2608.24007 · 2026-09-11超越 $w_0-w_a$ 与幻影穿越：检验暗物质与暗能量耦合模型
Beyond $w_0-w_a$ and Phantom Crossing: Testing Models of Coupled Dark Sector
arXiv 2608.23712 · 2026-09-11使用补充样本检验参数模型中可观测的选择问题
Testing selection on observables in parametric models with refreshment samples
arXiv 2608.23508 · 2026-09-11DelistBench：评估支持搜索的大型语言模型用于可审计的公司事件数据库补全
DelistBench: Evaluating Search-Enabled LLMs for Auditable Corporate-Event Database Completion
arXiv 2608.22770 · 2026-09-11
