2026-09论文中文摘要
基于置信分布传播的随机效应Meta分析的频率论预测区间
Frequentist prediction intervals for random-effects meta-analysis via confidence-distribution propagation
arXiv 2608.26527 · 2026-09-18缓解招聘领域多阶段大语言模型（LLM）流水线的生成幻觉：对提示护栏与人机交互（HITL）检查点的实证评估
Mitigating Fabrication in Multi-Stage LLM Pipelines for Hiring: An Empirical Evaluation of Prompt Guardrails and Human-in-the-Loop Checkpoints
arXiv 2608.26171 · 2026-09-18大语言模型查询模拟中的“知识诅咒”：用于追踪答案侧侵入的概念来源
The "Curse of Knowledge" in LLM Query Simulation: Concept Provenance for Tracing Answer-Side Intrusion
arXiv 2608.25245 · 2026-09-18D³-MOPD：面向高效多教师蒸馏的自适应动态域调度
D$^3$-MOPD: Dynamic Domain ScheDuling for Efficient Multi-Teacher Distillation
arXiv 2608.24987 · 2026-09-18PonderPounce：作为机器人控制回合上下文引擎的预训练多模态大语言模型（MLLM）
PonderPounce: A Pretrained MLLM as an Episode Context Engine for Robot Control
arXiv 2608.24115 · 2026-09-18动量-标量耦合湍流及其反常动量与标量扩散：第1部分——无外力与含长程外力
Momentum-scalar coupled turbulence with anomalous momentum and scalar diffusions. Part 1: Without external force and with long-range external force
arXiv 2608.23946 · 2026-09-18可表示拟阵的弗罗贝尼乌斯幂理想与超平面规避
A nowhere-zero point for several linear mappings simultaneously
arXiv 2608.23647 · 2026-09-18关于$D(G)^c$的分次中心
On the graded center of $D(G)^c$
arXiv 2608.23427 · 2026-09-18LongWoF-Bench：评估用于可验证长工作流任务的EvoMap基因
LongWoF-Bench: Evaluating EvoMap Genes for Verifiable Long-Workflow Tasks
arXiv 2608.23200 · 2026-09-18CatchBench：智能体故障何时能被检测？
CatchBench: When Can an Agent Failure Be Caught?
arXiv 2608.22808 · 2026-09-18稀缺确诊性PET测量的分配：A4/LEARN中的目标对齐验证
Spending Scarce Confirmatory PET Measurements: Target-Aligned Validation in A4/LEARN
arXiv 2608.22223 · 2026-09-18LMP-GNN：用于带符号最大压力交通信号控制的缺失车道流量概率重建
LMP-GNN: Probabilistic Reconstruction of Missing Lane Counts for Signed Max-Pressure Traffic Signal Control
arXiv 2608.21734 · 2026-09-18泡利概率谱携带纯态量子费希尔度量
The Pauli Probability Spectrum Carries the Pure-State Quantum Fisher Metric
arXiv 2608.21437 · 2026-09-18AIREP：AI运行时治理中逐决策证据的协议
AIREP: A Protocol for Per-Decision Evidence in AI Runtime Governance
arXiv 2608.21363 · 2026-09-18平丛典范扩张的推广陈-西蒙斯类的挠率
Torsion of extended Chern-Simons classes for canonical extensions of flat bundles
arXiv 2608.20877 · 2026-09-18通用均匀超图的调色板稀疏化
Streaming Hypergraph Coloring via Palette Sparsification
arXiv 2608.19623 · 2026-09-18俘获：Vlasov-Poisson动力学的“瀑布”以及无碰撞等离子体中与漂移无关的空结构的瞬态后出现
Trapping: The "Waterfall" of Vlasov-Poisson Dynamics and the Post-Transient Emergence of Drift-Independent Hole Structures in Collisionless Plasmas
arXiv 2608.18658 · 2026-09-18GigaBrain-WBC-0.5：一种用于与环境交互的鲁棒全身控制的行为世界模型
GigaBrain-WBC-0.5: A Behavior World Model for Robust Humanoid Whole-Body Tracking with Environment Interaction
arXiv 2608.18234 · 2026-09-18Putnam-Fuglede交换性与初等算子的值域-核正交性
Putnam-Fuglede commutativity and the range-kernel orthogonality of an elementary operator
arXiv 2608.18125 · 2026-09-18AutoResearch：输入洞见，输出无幻觉
AutoResearch: Insight In, Hallucination Out
arXiv 2608.17906 · 2026-09-18拟芬斯勒度量的等周问题与最小空气动力学阻力的形状
Isoperimetric problem for quasi-Finsler metrics and shapes of least aerodynamic resistance
arXiv 2608.17783 · 2026-09-18横向场伊辛模型在图学习中的表达能力研究
On the Expressive Power of the Transverse-Field Ising Model for Graph Learning
arXiv 2608.17750 · 2026-09-18开源统一瞬态多相先进反应堆模拟工具包（Outram Park）中库的智能体迁移、构建及初步验证与确认 第一部分：热工水力
Agentic Porting, Construction and Initial Verification and Validation of Libraries within the Open Source Unified TRAnsient Multi-Phase Advanced Reactor simulation Kit (Outram Park) Part I: A Software Engineering Log for Thermal Hydraulics
arXiv 2608.17504 · 2026-09-18简短公告：拜占庭SMR中隐藏状态授权的公平绑定
Brief Announcement: Fair Binding for Hidden-State Authorization in Byzantine SMR
arXiv 2608.17349 · 2026-09-18对称强制变分问题神经近似的无参考记录能量-神谕恢复：符合里泽重构与存档级选择
Reference-free logged energy-oracle recovery for neural approximations of symmetric coercive variational problems: conforming Riesz reconstruction and archive-level selection
arXiv 2608.16473 · 2026-09-18通过缪子对撞机的单光子加缺失能量信号探测类轴子粒子（ALP）-光子-暗光子相互作用
Probing the ALP--Photon--Dark Photon Interaction through the Mono-Photon plus Missing-Energy Signature at a Muon Collider
arXiv 2608.15656 · 2026-09-18学习序列出行选择：基于逆强化学习与模仿学习的路径与活动选择综述
Learning Sequential Mobility Choice: A Review of Route and Activity Choice through Inverse Reinforcement Learning and Imitation Learning
arXiv 2608.15339 · 2026-09-18受体信号网络中的幽灵动力学：竞争型癌症抑制模型的快慢自适应扩展
Ghost Dynamics in Receptor Signalling Networks: A Fast--Slow Adaptive Extension of Competitive Cancer Inhibition Models
arXiv 2608.15300 · 2026-09-18归因于光谱基础模型中的预处理不变性
Attributing Preprocessing Invariance in Spectral Foundation Models
arXiv 2608.14227 · 2026-09-18OccPlanner：面向像素目标导航的目标感知占用条件扩散规划器
OccPlanner: Goal-Aware Occupancy-Conditioned Diffusion Planner for PixelGoal Navigation
arXiv 2608.14160 · 2026-09-18基于专家对齐草稿的解耦对比解码
Decoupled Contrastive Decoding via Expert-Aligned Drafting
arXiv 2608.12913 · 2026-09-18ViTOED：面向越南社交媒体文本的定向情感检测数据集
ViTOED: A Dataset for Target-Oriented Emotion Detection on Vietnamese Social Media Texts
arXiv 2608.12776 · 2026-09-18SU(3)_F极限下Ω^0_c衰变的拓扑图
Topological diagrams of $Ω^0_c$ decays in the $SU(3)_F$ limit
arXiv 2608.12742 · 2026-09-18多组分等离子体大气中的双极电场：α粒子与随机加热的效应
The Ambipolar electric field in multispecies plasma atmospheres: effects of alpha particles and stochastic heating
arXiv 2608.12014 · 2026-09-18加密货币与股票市场中典型事实的普适性与异质性
Universality and Heterogeneity of Stylized Facts in Cryptocurrency and Equity Markets
arXiv 2608.10852 · 2026-09-18强耦合软函数
Strongly Coupled Soft Functions
arXiv 2608.10083 · 2026-09-18带边界曲面的Dirichlet-to-Neumann映射的行列式及其二重曲面上全纯微分的周期
The determinant of the Dirichlet-to-Neumann map for a surface with boundary and periods of holomorphic differentials on its double
arXiv 2608.09737 · 2026-09-18用于高带宽透明物理交互的超低阻抗机器人夹爪
Ultra-Low-Impedance Robotic Gripper for High-Bandwidth and Transparent Physical Interaction
arXiv 2608.09198 · 2026-09-18FitAQA：面向多模态大语言模型的健身动作质量评估基准
FitAQA: A Benchmark of Fitness Action Quality Assessment for Multimodal Large Language Models
arXiv 2608.08736 · 2026-09-18一维周期三次非线性薛定谔方程奇异解的强非唯一性
Sharp non-uniqueness of singular solutions to the one-dimensional periodic cubic nonlinear Schrödinger equation
arXiv 2608.07866 · 2026-09-18保持简洁：用于超长视频理解的多键情景记忆检索
Keep It Simple: Multi-Key Episodic Memory Retrieval for Ultra-Long Video Understanding
arXiv 2608.07663 · 2026-09-18能级统计下的解析传播复杂度：从混沌到可积性
Analytic Spread Complexity from Level Statistics: From Chaos to Integrability
arXiv 2608.07412 · 2026-09-18Radon数的超线性下界
On colorful Helly numbers and the growth of Tverberg numbers
arXiv 2608.06437 · 2026-09-18E7中的三代（费米子）
Three Generations in E7
arXiv 2608.06271 · 2026-09-18当自进化适得其反：针对LLM智能体中技能污染的预提交门控机制
When Self-Evolution Backfires: Pre-Commit Gating against Skill Contamination in LLM Agents
arXiv 2608.05810 · 2026-09-18有限并行访问下的量子信道学习
Quantum channel learning with limited parallel access
arXiv 2608.05307 · 2026-09-18关于随机矩阵的热流猜想
On the heat flow conjecture for random matrices
arXiv 2608.05185 · 2026-09-18用于视觉-语言-动作模型长程规划的显式语言记忆
Explicit Language Memory for Long-Horizon Planning in Vision-Language-Action Models
arXiv 2608.04765 · 2026-09-18基于分离器的图编辑距离问题算法
A Separator-based Algorithm for the Graph Edit Distance Problem
arXiv 2608.04583 · 2026-09-18CodeAssay：带有审计基准真值的多指标大语言模型代码生成基准
CodeAssay: A Multi-Metric Benchmark with Audited Ground Truth for LLM Code Generation
arXiv 2608.03535 · 2026-09-18
