2026-09论文中文摘要
切片Orlicz-Wasserstein距离
Sliced Orlicz-Wasserstein
arXiv 2609.32847 · 2026-09-29SynCo：通过对比交互残差学习跨模态协同
SynCo: Learning Cross-Modal Synergy by Contrasting Interaction Residuals
arXiv 2609.32846 · 2026-09-29腔-原子系统中利用三能级辅助粒子生成任意量子态
Generation of arbitrary quantum states in cavity-atom systems with a three-level ancilla
arXiv 2609.32845 · 2026-09-29Mask2Restore：基于修复的自监督超声去斑
Mask2Restore: Self-Supervised Ultrasound Despeckling via Inpainting
arXiv 2609.32844 · 2026-09-29WhisperVC-AV：用于噪声鲁棒的耳语到正常语音转换的音视频内容恢复
WhisperVC-AV: Audio-Visual Content Restoration for Noise-Robust Whisper-to-Normal Voice Conversion
arXiv 2609.32843 · 2026-09-29具有动态阈值的混合神经元模型中的阈下振荡与放电
Subthreshold oscillations and spiking in hybrid neuron models with a dynamic threshold
arXiv 2609.32842 · 2026-09-29超越时间平滑：空间能量预算稳定一步扩散编辑
Beyond Temporal Smoothing: Spatial Energy Budgets Stabilize One-Step Diffusion Editing
arXiv 2609.32841 · 2026-09-29VCRE-Fib：基于视图条件区域证据的日本血吸虫相关肝纤维化细粒度超声分级
VCRE-Fib: View-Conditioned Regional Evidence for Fine-Grained Ultrasound Grading of Schistosoma japonicum-Associated Liver Fibrosis
arXiv 2609.32840 · 2026-09-29弥合测量差距：面向短视频推荐的潜在用户偏好建模
Mend the Measurement Gap: Latent User Preference Modeling for Short-Form Video Recommendation
arXiv 2609.32839 · 2026-09-29HamiFormer：具有仿射辛映射的双专家扩散场
HamiFormer: Dual-Expert Diffusion Fields with Affine Symplectic Maps
arXiv 2609.32838 · 2026-09-29边扫描边想象：用于机器人超声导航的场景图世界模型
Scanning While Imagining: A Scene-Graph World Model for Robotic Ultrasound Navigation
arXiv 2609.32837 · 2026-09-29不完整观测下的无线电图盲预测：误差表征与可修正传播先验学习
Radiomap Blind Prediction under Incomplete Observation: Error Characterization and Correctable Propagation-Prior Learning
arXiv 2609.32836 · 2026-09-29FinancialAuditBench：基于真实世界先验的差分隐私基准构建
FinancialAuditBench: Benchmark Construction under Differential Privacy Using Real-World Priors
arXiv 2609.32835 · 2026-09-29代数几何Parvaresh--Vardy子空间设计与秩凝聚器
Algebraic-Geometric Parvaresh--Vardy Subspace Designs and Rank Condensers
arXiv 2609.32834 · 2026-09-29置换等变流匹配用于免对齐神经权重生成
Permutation-Equivariant Flow Matching for Alignment-Free Neural Weight Generation
arXiv 2609.32833 · 2026-09-29异构移动无线网络中的空中联邦学习
Over-the-Air Federated Learning in Heterogeneous Mobile Wireless Networks
arXiv 2609.32832 · 2026-09-29UniCache：面向统一多模态模型的任务与类型感知KV缓存压缩
UniCache: Task- and Type-Aware KV Cache Compression for Unified Multimodal Models
arXiv 2609.32831 · 2026-09-29关于预李代数中的幂零性
On nilpotency in pre - Lie algebras
arXiv 2609.32830 · 2026-09-29具有齐次Neumann边界条件的Allen-Cahn方程过渡层的二阶估计：第一部分——边界正交过渡层
Second order estimates on transition layers for the Allen-Cahn equation with homogeneous Neumann boundary conditions: Part I - Boundary-orthogonal transition layers
arXiv 2609.32829 · 2026-09-29重定时贝尔曼流：逃离速度自举的不可能三角
Retimed Bellman Flows: Escaping the Impossible Triangle of Velocity Bootstrapping
arXiv 2609.32828 · 2026-09-29通过多智能体偏好学习改进LLM协作
Improving LLM Collaboration via Multi-Agent Preference Learning
arXiv 2609.32827 · 2026-09-29单晶硅鳍状势垒的数字刻蚀降至亚10纳米
Digital Etching of Single-Crystalline Silicon Fin Barriers Down to Sub-10 nm
arXiv 2609.32826 · 2026-09-29分解税：LLM流水线在其自身接口处损失高达40个准确率点
The Decomposition Tax: LLM Pipelines Lose Up to 40 Accuracy Points at Their Own Interfaces
arXiv 2609.32825 · 2026-09-29Social3Source：连接开源软件基金会中的社会溯源、社会品质与社会目的
Social3Source: Connecting Social Provenance, Societal Qualities, and Social Purpose on an Open-Source Software Foundation
arXiv 2609.32823 · 2026-09-29多输入多输出雷达中混合分集波形方案效应的表征
Characterizing the Effects of Mixed Division Waveform Schemes in MIMO Radar
arXiv 2609.32822 · 2026-09-29路由漂移本身不足以诊断合并MoE大语言模型中的故障
Routing Drift Alone Does Not Diagnose Failure in Merged MoE LLMs
arXiv 2609.32821 · 2026-09-29使用成员查询和对比示例学习洗牌理想
Learning Shuffle Ideals with Membership Queries and Contrastive Examples
arXiv 2609.32820 · 2026-09-29秩坍缩可恢复，而$|Q|$增长不可恢复：高UTD Soft Actor-Critic中价值发散的超前样本外预警
Rank Collapse Is Recoverable, Growing $|Q|$ Is Not: Out-of-Sample Early Warning for Value Divergence in High-UTD Soft Actor-Critic
arXiv 2609.32819 · 2026-09-29何时微调的一阶模型能够界定遗忘？
When Can First-Order Models of Fine-Tuning Bound Forgetting?
arXiv 2609.32818 · 2026-09-29正确答案，错误理由：准确性、一致性和共识是临床决策支持中LLM忠实度的误导性指标
Right Answer, Wrong Reason: Accuracy, Consistency, and Consensus Are Misleading Indicators of LLM Faithfulness in Clinical Decision Support
arXiv 2609.32817 · 2026-09-29集合的第三次对称积在 HoTT 中是一个集合
The third symmetric product of a set is a set in HoTT
arXiv 2609.32816 · 2026-09-29倒数平衡数立方和的无穷级数
The infinite sum of the cubes of reciprocal balancing numbers
arXiv 2609.32815 · 2026-09-29改变乘积，保持参数：Transformer的关联代数层
Change the Product, Keep the Parameters: Associative Algebra Layers for Transformers
arXiv 2609.32814 · 2026-09-29USAI-Quant：建筑环境中视觉语言模型的定量推理基准
USAI-Quant: A Quantitative Reasoning Benchmark for Vision-Language Models in Built Environments
arXiv 2609.32813 · 2026-09-29具有变号系数的Liouville系统的同步解
Synchronized Solutions for Liouville Systems with Sign-Changing Coefficients
arXiv 2609.32812 · 2026-09-29渐进式风险评估用于事故预判
Progressive Risk Estimation for Accident Anticipation
arXiv 2609.32811 · 2026-09-29选择过多：研究大规模下LLM的决策制定
Overwhelmed by Choice: Studying LLM Decision Making at Scale
arXiv 2609.32809 · 2026-09-29注意尖峰：大型视觉-语言模型中视觉大规模激活的机制与脆弱性
Mind the Spike: Mechanisms and Brittleness of Visual Massive Activations in Large Vision-Language Models
arXiv 2609.32808 · 2026-09-29超越准确性：临床评估大语言模型中的反事实脆弱性与人口统计偏见
Beyond Accuracy: Counterfactual Fragility and Demographic Bias in Clinical Evaluation of LLMs
arXiv 2609.32807 · 2026-09-29寻找情感所属之处：通过掩码时间情感锚定重新思考音频情感识别
Finding Emotions Where They Belong: Rethinking Audio Emotion Recognition through Masked Temporal Affective Grounding
arXiv 2609.32804 · 2026-09-29Nutri-ATLAS：面向更智能营养的表格查询与辅助的具身智能体
Nutri-ATLAS: Embodied Agent for Tabulated Lookup and Assistance for Smarter nutrition
arXiv 2609.32803 · 2026-09-29可重推导性决定分阶段智能体流水线在上游故障后恢复什么
Re-derivability Decides What a Staged Agent Pipeline Recovers After an Upstream Fault
arXiv 2609.32802 · 2026-09-29PlanGuard：具身智能体中多步计划安全性的防护栏
PlanGuard: A Guardrail for Multi-Step Plan Safety in Embodied Agents
arXiv 2609.32801 · 2026-09-29电子-离子对撞机上的量子层析
Quantum tomography at the Electron-Ion Collider
arXiv 2609.32800 · 2026-09-29并列情形下的奇偶检验：一个单检验提升定理
Parity Tests under Ties: A One-Test Lifting Theorem
arXiv 2609.32799 · 2026-09-29自适应与弹性双层资源切片用于悬停空中回程网络
Adaptive and Resilient Dual-Layer Resource Slicing for Hovering Aerial Backhaul Networks
arXiv 2609.32798 · 2026-09-29通过草图切线一致性进行神经算子的导数信息在线训练
Derivative-Informed Training of Neural Operators On-the-Fly via Sketched Tangent Consistency
arXiv 2609.32797 · 2026-09-29ERP-FM：面向通用ERP分析的基础模型
ERP-FM: A Foundation Model for Universal ERP Analysis
arXiv 2609.32796 · 2026-09-29AgentHabit：刻画智能体在日常任务中的不同行为
AgentHabit: Characterizing Distinct Behaviors of Agents on Everyday Tasks
arXiv 2609.32795 · 2026-09-29潜空间并非平坦：重新思考三维医学图像合成的潜结构
Latent Space Is Not Flat: Rethinking Latent Structure for 3D Medical Image Synthesis
arXiv 2609.32794 · 2026-09-29
