2026-09论文中文摘要
滚动时域控制与耗散性——最优控制、博弈与不确定性
Receding Horizon Control and Dissipativity - Optimal Control, Games and Uncertainty
arXiv 2609.36858 · 2026-09-30LUMA：用于天文成像中强引力透镜搜索的卷积神经网络
LUMA: A CNN for Strong Gravitational Lens Searches in Astronomical Imaging
arXiv 2609.36857 · 2026-09-30FinRegQA-EU：基于腐败偏好数据的欧盟金融监管问答
FinRegQA-EU: Corruption-Based Preference Data for Grounded EU Financial Regulatory Question Answering
arXiv 2609.36856 · 2026-09-30当上游消息覆盖正确答案：多智能体LLM协作的受控研究
When Upstream Messages Override Correct Answers: A Controlled Study of Multi-Agent LLM Collaboration
arXiv 2609.36855 · 2026-09-30多元信号恢复：基于多折叠图学习
Multivariate signal restoration via Multifold Graph Learning
arXiv 2609.36854 · 2026-09-30爱因斯坦-卡特自旋宇宙学中的全息暗能量
Holographic dark energy in Einstein-Cartan spin cosmology
arXiv 2609.36853 · 2026-09-30社会性锚点：面向群体有界轨迹预测
Socialality Anchors: Towards Group-bounded Trajectory Prediction
arXiv 2609.36852 · 2026-09-30重新思考生成式人工智能时代的多模态假新闻检测
Rethinking Multimodal Fake News Detection in the Generative AI Era
arXiv 2609.36850 · 2026-09-30不安全的梯度能否在对话中幸存？多轮对话中基于梯度的越狱检测的脆弱性研究
Does the Unsafe Gradient Survive a Conversation? On the Fragility of Gradient-Based Jailbreak Detection in Multi-Turn Dialogue
arXiv 2609.36849 · 2026-09-30一种有效、可靠且稳健的基于可穿戴传感器的人类活动识别框架
An Effective, Reliable, and Robust Framework for Human Activity Recognition Using Wearable Sensors
arXiv 2609.36848 · 2026-09-30基于统计形状模型和深度学习的骨盆骨折复位螺钉自动规划
Automated Screw Planning for Reduced Pelvic Fractures Based on Statistical Shape Models and Deep Learning
arXiv 2609.36847 · 2026-09-30耦合无序系统中的锁定转变
Locking transition in coupled disordered systems
arXiv 2609.36846 · 2026-09-30DSWM：面向需求驱动无人机基站重定位的分解时空世界模型
DSWM: Decomposed Spatio-Temporal World Model for Demand-Driven UAV Base Station Repositioning
arXiv 2609.36845 · 2026-09-30GlassFormer：利用雷达-深度融合实现实时玻璃分割
GlassFormer: Learning Real-time Glass Segmentation using Radar-Depth Fusion
arXiv 2609.36844 · 2026-09-30RolloutFaith：审计视觉世界模型中的持久内部干预
RolloutFaith: Auditing Persistent Internal Interventions in Visual World Model
arXiv 2609.36843 · 2026-09-30VGGT 家族是否需要其所有层？
Does the VGGT Family Need All Its Layers?
arXiv 2609.36842 · 2026-09-30辐射缪子衰变的算符分析
An Operator Analysis of Radiative Muon Decay
arXiv 2609.36841 · 2026-09-30SDRE模型的高性能计算与离散卡尔曼滤波用于鲁棒$H_{\infty}$控制
High Performance Computing of SDRE models with Discrete Kalman Filtering for Robust $H_{\infty}$-Controls
arXiv 2609.36840 · 2026-09-30BRP-GABLE：一种面向多种结局类型的可解释基于规则的预测框架
BRP-GABLE: An Interpretable Rule-Based Prediction Framework for Multiple Outcome Types
arXiv 2609.36839 · 2026-09-30基于策略的视觉证据蒸馏
On-Policy Visual Evidence Distillation
arXiv 2609.36838 · 2026-09-30你无法恢复从未被测量的信息：量化超低场MRI超分辨率的信息上限
You Cannot Recover What Was Never Measured: Quantifying the Information Ceiling of Ultra-Low-Field MRI Super-Resolution
arXiv 2609.36837 · 2026-09-30测量-反馈间隔控制量子玻尔兹曼采样器中的有序化
Ordering controlled by the measurement-feedback interval in quantum Boltzmann samplers
arXiv 2609.36836 · 2026-09-30ARC-KV：基于重建的KV缓存压缩中的锚点搜索摊销
ARC-KV: Amortizing Anchor Search for Reconstruction-Based KV Cache Compaction
arXiv 2609.36835 · 2026-09-30WenetSpeech-Min：用于方言语音处理的大规模闽南语语音语料库及双转录
WenetSpeech-Min: A Large-Scale Minnan Speech Corpus with Dual Transcriptions for Dialectal Speech Processing
arXiv 2609.36834 · 2026-09-30MHD湍流中动力学状态的形态学识别：ScaleAware-JEPA
Morphological Identification of Dynamical Regimes in MHD Turbulence with ScaleAware-JEPA
arXiv 2609.36833 · 2026-09-30视频扩散模型中的运动概念遗忘
Motion Concept Unlearning in Video Diffusion Models
arXiv 2609.36832 · 2026-09-30Oldroyd-B 流体径向流动：超稀极限下的理论与模拟结果
Radial flow of an Oldroyd-B fluid: theoretical and simulation results in the ultra-dilute limit
arXiv 2609.36831 · 2026-09-30陈旧性在哪里累积？面向LLM后训练中异步RL的池感知有效陈旧性控制
Where Does Staleness Accumulate? Pool Aware Effective Staleness Control for Asynchronous RL in LLM Post-Training
arXiv 2609.36830 · 2026-09-30默认陷阱：重新思考工具使用型LLM智能体中的计划评估
The Default Trap: Rethinking Plan Evaluation in Tool-Using LLM Agents
arXiv 2609.36829 · 2026-09-30校准改变未来的决策：多模态大语言模型的在线策略后训练量化
Calibrate the Decisions That Change the Future: On-Policy Post-Training Quantization for Multimodal Large Language Models
arXiv 2609.36828 · 2026-09-30最优Max-Cut SDP解的几何结构
The Geometry of Optimal Max-Cut SDP Solutions
arXiv 2609.36827 · 2026-09-30基于自一致性的扩散视频推理学习
Learning via Self-Consistency for Diffusion-based Video Reasoning
arXiv 2609.36826 · 2026-09-30COVID-19旅游冲击期间奢侈品转售量和构成发生变化，但共同购买网络检验结果不确定且对构建方式敏感
Luxury resale volume and composition shifted during the COVID-19 tourism shock, but co-purchase network tests were inconclusive and construction-sensitive
arXiv 2609.36825 · 2026-09-30检验删失持续时间模型中的未观测异质性：EM方法
Testing for Unobserved Heterogeneity in Censored Duration Models: EM Approach
arXiv 2609.36824 · 2026-09-30层级群的模复形
Complexes of modules over hierarchical groups
arXiv 2609.36823 · 2026-09-30RED：面向泛化AI生成图像检测的重建演化动力学
RED: Reconstruction Evolution Dynamics for Generalizable AI-Generated Image Detection
arXiv 2609.36822 · 2026-09-30半线性热方程无穷多个奇异自相似解的存在性
Existence of infinitely many singular self-similar solutions for the semilinear heat equation
arXiv 2609.36821 · 2026-09-30CorrGRPO：用于多奖励学习的相关性归一化GRPO
CorrGRPO: Correlation-Normalized GRPO for Multi-Reward Learning
arXiv 2609.36820 · 2026-09-30固定阶线性形变中 Milnor 数的跳跃
Jump of the Milnor Number in Linear Deformations with Fixed Order
arXiv 2609.36819 · 2026-09-30pikit：用于间接提示注入研究与评估的可组合工具包
pikit: A Composable Toolkit for Indirect Prompt Injection Research and Evaluation
arXiv 2609.36817 · 2026-09-30迈向更好的训练信号：优势裁剪策略优化
Towards Better Training Signal: Advantage Clipped Policy Optimization
arXiv 2609.36816 · 2026-09-30CurvSpec：面向部分相关视频检索的自适应多曲率学习
CurvSpec: Adaptive Multi-Curvature Learning for Partial Relevant Video Retrieval
arXiv 2609.36815 · 2026-09-30RESCUE：通过强化学习将语言模型错误修复为稀疏电路
RESCUE: Repairing Language Model Errors to Sparse Circuits via Reinforcement Learning
arXiv 2609.36813 · 2026-09-30单生成性的一种局部方法及其在 Lenny Jones 猜想中的应用
A Local Approach to Monogenity with an Application to Lenny Jones' Conjecture
arXiv 2609.36811 · 2026-09-30MeteoVerse：统一的可控天气视频世界模型
MeteoVerse: Unified Weather-Controllable Video World Model
arXiv 2609.36810 · 2026-09-30几何条件化的固定支架编码器用于时间扭曲鲁棒的序列检索
Geometry-Conditioned Fixed-Scaffold Encoders for Time-Warp Robust Sequence Retrieval
arXiv 2609.36809 · 2026-09-30Spotter：让具身模型主导，VLM 反思
Spotter: Let the Embodied Model Lead, and the VLM Reflect for It
arXiv 2609.36808 · 2026-09-30XRepoSkill：为软件工程智能体学习可迁移技能
XRepoSkill: Learning Transferable Skills for Software Engineering Agents
arXiv 2609.36807 · 2026-09-30UpliftMem：学习智能体记忆检索的集合级提升
UpliftMem: Learning Set-Level Uplift for Agent Memory Retrieval
arXiv 2609.36805 · 2026-09-30VAA-CSEC：用于中文语义纠错的投票引导优势分配
VAA-CSEC: Vote-guided Advantage Allocation for Chinese Semantic Error Correction
arXiv 2609.36804 · 2026-09-30
