2026-09论文中文摘要
用于再电离时期CMB可观测量的自洽、物理且解析的模型 I. 自谱的推导、约束能力与可探测性
A consistent, physical, and analytical model for CMB observables of reionisation I. Derivation, constraining power, and detectability of the auto-spectra
arXiv 2608.30766 · 2026-09-01T3S：利用任务特定特征选择器与调度器改进多任务强化学习
T3S: Improving Multi-Task Reinforcement Learning with Task-Specific Feature Selector and Scheduler
arXiv 2608.30765 · 2026-09-01特殊断点图的循环结构生成函数
Cycle-Structure Generating Functions for Special Breakpoint Graphs
arXiv 2608.30764 · 2026-09-01通过正特征函数的双侧界证明带非局部Robin边界条件的热半群的超收缩性
Ultracontractivity of heat semigroups with non-local Robin boundary conditions via two-sided bounds for a positive eigenfunction
arXiv 2608.30763 · 2026-09-01ChessQueries：迈向更优的国际象棋棋盘识别
ChessQueries: Toward Better Chess Board Recognition
arXiv 2608.30762 · 2026-09-01缓变相互作用链中的共振统计、福克空间分支与长程对网络
Resonance statistics, Fock-space branching, and long-range pair networks in slowly varying interacting chains
arXiv 2608.30761 · 2026-09-01PRACTICE：从经验到自进化具身智能体的专业能力
PRACTICE: From Experience to Expertise in Self-Evolving Embodied Agents
arXiv 2608.30760 · 2026-09-01极端视界固有刚性的几何起源
Geometric origin of intrinsic rigidity for extremal horizons
arXiv 2608.30759 · 2026-09-01CheXGround：用于 grounded 纵向胸部 X 线片解读的解剖区域标记
CheXGround: Anatomical Region Tokens for Grounded Longitudinal Chest X-ray Interpretation
arXiv 2608.30758 · 2026-09-01哪些规则现在重要？智能系统行动前的策略质心路由
Which Rules Matter Now? Policy-Centroid Routing Before an Intelligent System Acts
arXiv 2608.30757 · 2026-09-01动态大语言模型（LLM）对话在软件开发中的应用前景
On the Prospects of Dynamic LLM Conversations in Software Development
arXiv 2608.30756 · 2026-09-01克尔-贝托蒂-罗宾逊时空中偶极粒子轨道与临界碰撞的外磁场效应
External magnetic-field effects on dipolar-particle orbits and critical collisions in Kerr--Bertotti--Robinson spacetime
arXiv 2608.30755 · 2026-09-01CLIN：一种用于评估波斯语短篇文学文本创造性的客观框架
CLIN: an Objective Framework for Evaluating Creativity in Short Persian Literary Text
arXiv 2608.30754 · 2026-09-01从检索内容中学习：面向语义检索的在线强化学习微调
Learning from What You Retrieve: Online RL Fine-Tuning for Semantic Retrieval
arXiv 2608.30753 · 2026-09-01Teichmüller测地线沿能量函数的拟凸性
Quasi-convexity of energy functions along Teichmüller geodesics
arXiv 2608.30752 · 2026-09-01视觉语言模型（VLMs）是否在不同模态间共享安全神经元？
Do VLMs Share Safety Neurons Across Modalities?
arXiv 2608.30750 · 2026-09-01结合COS方法的重要性采样在投资组合风险分配中的重要性
Importance Sampling Enhanced with the COS Method for the Portfolio Risk Allocation
arXiv 2608.30749 · 2026-09-01不同运行状态下越狱鲁棒性的脆弱性
The Fragility of Jailbreak Robustness Across Operational States
arXiv 2608.30748 · 2026-09-01非周期链中的哈特里碎裂与长程对网络
Hartree Fragmentation and Long-Range Pair Networks in Aperiodic Chains
arXiv 2608.30746 · 2026-09-01非线性平面迭代函数系统（IFS）的表面排斥子与吸引子的维数
Dimensions of surface repellers and attractors of non-linear planar IFSs
arXiv 2608.30744 · 2026-09-01关于李群胚的Bott谱序列的若干观点
Perspectives on the Bott spectral sequence for Lie groupoids
arXiv 2608.30743 · 2026-09-01拓扑异构酶II存在下环形纳米通道中dsDNA多连环体与网络形成的粗粒度模拟
Coarse-grained simulations of dsDNA polycatenanes and network formation in annular nanochannels with topoisomerase II
arXiv 2608.30742 · 2026-09-01神经网络中的功能退化：测量与剪枝
Functional Degeneracy in Neural Networks: Measurement and Pruning
arXiv 2608.30741 · 2026-09-01高阶稳定dg范畴与簇Morita理论的Auslander对应
Auslander correspondence for higher stable dg categories and cluster Morita theory
arXiv 2608.30740 · 2026-09-01并非所有弃权（不执行）都是失败：理解与恢复移动语音助手中的弃权（不执行）情况
Not All Fallbacks Are Failures: Understanding and Recovering from Fallbacks in Mobile Voice Assistants
arXiv 2608.30738 · 2026-09-01SU(3)味对称点下的 $D \to (K π)_{\mathbf{27}}$ 衰变：第一部分：方法论与强相位确定
$D \to (K π)_{\mathbf{27}}$ at the SU(3)-flavour-symmetric point I: Methodology and strong phase determination
arXiv 2608.30737 · 2026-09-01依赖分布的随机Volterra方程的稳定性结果
Stability results for distribution-dependent stochastic Volterra equations
arXiv 2608.30736 · 2026-09-01√Δ-精细环
$\sqrtΔ$-Fine Rings
arXiv 2608.30734 · 2026-09-01梯度折射率多模光纤中的鲁棒双稳态涡旋光子弹
Robust bistable vortex light bullets in graded-index multimode fibers
arXiv 2608.30732 · 2026-09-01电商基准测试集：评估大语言模型智能体在长周期自主商业运营中的表现
E-Commerce Bench: Evaluating LLM Agents on Long-Horizon Autonomous Business Operation
arXiv 2608.30730 · 2026-09-01可编程散射系统调控部分相干微波的界限
Bounds on Shaping Partially Coherent Microwaves with Programmable Scattering Systems
arXiv 2608.30729 · 2026-09-01关于若干几何图凸性的最大权凸问题
On the maximum weight convex problem for some geometric graph-convexities
arXiv 2608.30728 · 2026-09-01RailGen：通过智能体引导的小规模异物生成改进铁路入侵检测
RailGen: Improving Railway Intrusion Detection via Agent-Guided Small-Scale Foreign Object Generation
arXiv 2608.30727 · 2026-09-01面向情感分析的、结合文本路由与序数原型优化的多模态自适应专家选择方法
Multimodal Adaptive Expert Selection with Text Routing and Ordinal Prototype Optimization for Sentiment Analysis
arXiv 2608.30726 · 2026-09-01多模态视觉-语言编码器在低资源语言上的失效位置
Where Do Multilingual Vision-Language Encoders Fail on Low-Resource Languages?
arXiv 2608.30725 · 2026-09-01BAITBENCH：通过植入ML任务中的可选捷径测量智能体的奖励黑客行为
BAITBENCH: Measuring Agent Reward Hacking with Optional Shortcuts Planted in ML Tasks
arXiv 2608.30724 · 2026-09-01二阶矩与反集中假设下随机矩阵的中间奇异值
Intermediate Singular Values of Random Matrices under Second-Moment and Anti-Concentration Assumptions
arXiv 2608.30722 · 2026-09-01高对比度介质中的本征值渐近行为
Eigenvalue Asymptotics in High-Contrast Media
arXiv 2608.30721 · 2026-09-01通过带随机LayerNorm的Transformer处理追踪可区分性
Tracing distinguishability through transformer processing with stochastic LayerNorm
arXiv 2608.30720 · 2026-09-01一种可移动里德堡原子量子接收机的光路方案
An Optical Pathway to Movable Rydberg Atomic Quantum Receivers
arXiv 2608.30718 · 2026-09-01SocialReasonBench：基于反事实叙事视频的社会推理视频问答基准
SocialReasonBench: A Video-QA Benchmark for Social Reasoning with Counterfactual Narrative Videos
arXiv 2608.30716 · 2026-09-01端点能量原子在三维纳维-斯托克斯流中迫使局部压力集中
Endpoint Energy Atoms Force Local Pressure Concentration in Three-Dimensional Navier-Stokes Flow
arXiv 2608.30715 · 2026-09-01SegWave：基于小波的篡改区域分割
SegWave: Wavelet-Driven Segmentation of Tampered Regions
arXiv 2608.30714 · 2026-09-01闭合验证循环：用于长段落详细音频字幕的自检查字幕生成
Closing the Verification Loop: Self-Check Captioning for Long-Paragraph Detailed Audio Captioning
arXiv 2608.30713 · 2026-09-01GUIDE：用语言指令引导内部证据
GUIDE: Guiding Internal Evidence with Language Instructions
arXiv 2608.30712 · 2026-09-01带电星际尘埃的磁流体力学：多流体模型与线性模式研究
Magnetohydrodynamics of charged interstellar dust. Multifluid models and study of the linear modes
arXiv 2608.30711 · 2026-09-01柯尔莫哥洛夫-阿诺尔德对抗有界平移
Kolmogorov--Arnold against bounded translations
arXiv 2608.30710 · 2026-09-01RailSyn：面向铁路异物检测中可追溯数据补全的诊断引导图像生成方法
RailSyn: Diagnosis-Guided Image Generation for Traceable Data Completion in Railway Foreign Object Detection
arXiv 2608.30709 · 2026-09-01sbom-unifier：异构软件物料清单（SBOM）的集成框架
sbom-unifier: Integration Framework for Heterogeneous SBOMs
arXiv 2608.30708 · 2026-09-01伞状效应：磁冠拓扑结构调制空间平均的色球三分钟功率
The umbrella effect: Magnetic canopy topology modulates spatially averaged chromospheric three-minute power
arXiv 2608.30707 · 2026-09-01
