2026-09论文中文摘要
矩阵博弈中带赌博反馈的对数多项式纳什遗憾
Polylogarithmic Nash Regret in Matrix Games with Bandit Feedback
arXiv 2609.34812 · 2026-09-29UniOPSD：统一结果与事后反馈的智能体强化学习
UniOPSD: Unifying Outcome and Hindsight Feedback for Agentic Reinforcement Learning
arXiv 2609.34810 · 2026-09-29从感知到整合：重新审视视觉-语言模型推理的内部动态
From Perception to Integration: Revisiting the Internal Dynamics of Reasoning in Vision-Language Models
arXiv 2609.34809 · 2026-09-29HiCoMAC：基于计算导向调制设计的直方图状态编码多址计算
HiCoMAC: Histogram-State Coded Multiple Access Computing via Computation-Oriented Modulation Design
arXiv 2609.34808 · 2026-09-29ControlTrace：恢复控制场以实现隐藏内容识别
ControlTrace: Recovering Control Fields for Hidden-Content Recognition
arXiv 2609.34807 · 2026-09-29论大型音频语言模型中的时间绑定
On Temporal Binding in Large Audio Language Models
arXiv 2609.34806 · 2026-09-29SIPO：面向树结构智能体强化学习的选择性推断策略优化
SIPO: Selective-Inference Policy Optimization for Tree-Structured Agentic RL
arXiv 2609.34805 · 2026-09-29物理认证的联邦学习：保障关键水基础设施中协作异常检测的安全性
Physics-Attested Federated Learning: Securing Collaborative Anomaly Detection in Critical Water Infrastructure
arXiv 2609.34804 · 2026-09-29高斯视角下的通用光子减法
General Photon Subtraction from a Gaussian Perspective
arXiv 2609.34803 · 2026-09-29分离新用户校准脑电基础模型中的个人增益与群体增益
Separating personal from population gains when calibrating EEG foundation models for new users
arXiv 2609.34801 · 2026-09-29通过还是失败？评估大语言模型在两个希腊考试基准上的表现
Pass or Fail? Evaluating LLMs on Two Greek Examination Benchmarks
arXiv 2609.34800 · 2026-09-29跨房间被动WiFi跟踪：目标轨迹与虚拟发射器位置的联合估计
Cross-Room Passive WiFi Tracking: Joint Estimation of Target Trajectory and Virtual Transmitter Location
arXiv 2609.34797 · 2026-09-29硅氮氧化物相的组成：一项全面的热力学建模研究
Phase compositions of Si oxynitride: A comprehensive thermodynamic modeling study
arXiv 2609.34796 · 2026-09-29物理引导的光谱蒸馏用于资源受限设备上的水下图像增强
Physics-Guided Spectral Distillation for Underwater Image Enhancement on Resource-Constrained Devices
arXiv 2609.34795 · 2026-09-29混合连续变量与离散变量量子电路的高斯分解高效模拟
Efficient Simulation of Hybrid Continuous- and Discrete-Variable Quantum Circuits via Gaussian Decompositions
arXiv 2609.34793 · 2026-09-29带马尔可夫噪声的投影双时间尺度随机逼近的有限时间集中性与收敛速率
Finite-Time Concentration and Convergence Rates for Projected Two-Time-Scale Stochastic Approximation with Markov Noise
arXiv 2609.34791 · 2026-09-29非最小耦合引力的分离宇宙方法
The separate universe approach with non minimal coupling to gravity
arXiv 2609.34788 · 2026-09-29具有共形Maslov形式和勒让德毛细边界的拉格朗日子流形的刚性
Rigidity of Lagrangian submanifolds with conformal Maslov form and Legendrian capillary boundary
arXiv 2609.34787 · 2026-09-29实例自适应提示作为时间序列基础模型的上下文
Instance-Adaptive Prompts as Context for Time-Series Foundation Models
arXiv 2609.34786 · 2026-09-29BEHAVE：功能行为建模使硬件设计与验证智能体能够自我改进
BEHAVE: Functional Behavior Modeling Enables Self-Improving Agents for Hardware Design and Verification
arXiv 2609.34785 · 2026-09-29投影法向场：一种构造光滑UDF的凸优化方法
Projective Normal Fields: A Convex Optimization Method for Constructing Smooth UDFs
arXiv 2609.34784 · 2026-09-29TQTS-Bench：面向时间序列数据库的文本到查询多语法基准
TQTS-Bench: A Multi-Syntax Benchmark for Text-to-Query over Time-Series Databases
arXiv 2609.34783 · 2026-09-29CoHuB：一个多仿人机器人协作的仿真基准
CoHuB: A Simulation Benchmark for Multi-Humanoid Collaboration
arXiv 2609.34782 · 2026-09-29当视觉语言模型信任上下文：评估误导性上下文下的场景文本识别
When VLMs Trust Context: Evaluating Scene Text Recognition under Misleading Context
arXiv 2609.34781 · 2026-09-29并行信道与有记忆信道的散度最小化分布匹配
Divergence-Minimizing Distribution Matching for Parallel Channels and Channels with Memory
arXiv 2609.34779 · 2026-09-29TraceLib：用于语言无关Web模糊测试的系统调用位图反馈机制
TraceLib: System-Call Bitmap Feedback Mechanism for Language-Agnostic Web Fuzzing
arXiv 2609.34778 · 2026-09-29面向EvalLLM 2026的页面感知检索增强生成：关于法语PDF的五变体研究
Page-Aware Retrieval-Augmented Generation for EvalLLM 2026: A Five-Variant Study on French PDFs
arXiv 2609.34776 · 2026-09-29点确定性的可识别性
Recognizability of Being Point Determining
arXiv 2609.34775 · 2026-09-29三次相位门与囚禁离子振子
Cubic Phase Gate with a Trapped Ion Oscillator
arXiv 2609.34774 · 2026-09-29从对偶几何视角加速的Bregman近端梯度方法
Accelerated Bregman Proximal Gradient Methods from Dual Geometric Perspectives
arXiv 2609.34773 · 2026-09-29在令牌提交之前：扩散视觉语言模型中视觉幻觉的轨迹级基准测试
Before the Token Commits: Trajectory-Level Benchmarking of Visual Hallucinations in Diffusion VLMs
arXiv 2609.34772 · 2026-09-29模型内部何时有用？探索表示工程在LLM安全中的作用
When Do Model Internals Help? Exploring the Role of Representation Engineering in LLM Safety
arXiv 2609.34771 · 2026-09-29LongPuzzleBench：评估GUI智能体在长时程视觉谜题上的表现
LongPuzzleBench: Evaluating GUI Agents on Long-Horizon Visual Puzzles
arXiv 2609.34769 · 2026-09-29幂律分布耦合紧束缚模型中的局域化
Localization in tight-binding models with power-law distributed couplings
arXiv 2609.34767 · 2026-09-29关于 $\mathbb R^3$ 中静磁相互作用的注记
Notes on Magnetostatic Interactions in $\mathbb R^3$
arXiv 2609.34766 · 2026-09-29超越后训练量化中的重建损失：大型视觉-语言模型的平衡拟合
Beyond Reconstruction Loss in Post-Training Quantization: Balanced Fitting for Large Vision-Language Models
arXiv 2609.34765 · 2026-09-29WeaveData：具有自我批评与自我进化LLM计划的多模态数据分析系统
WeaveData: A Multimodal Data Analysis System with Self-Critiquing and Self-Evolving LLM Plans
arXiv 2609.34764 · 2026-09-29GPU加速的OTFS大规模MIMO在信道状态信息不确定性下的鲁棒波束成形
GPU-Accelerated Robust Beamforming for OTFS Massive MIMO with CSI Uncertainty
arXiv 2609.34763 · 2026-09-29应对冷链行业中的数据工程挑战，通过分析手段减少其环境影响
Addressing Data Engineering Challenges in the Cold-Chain Sector to Reduce its Environmental Impact Through Analytics
arXiv 2609.34762 · 2026-09-29基于模型势估算极性分子中正电子束缚态
Model potential based estimation of positron bound states with polar molecules
arXiv 2609.34761 · 2026-09-29关于空间 $BV^s(\R^n)$ 的简短注记
A short note about the spaces $BV^s(\R^n)$
arXiv 2609.34760 · 2026-09-29PanoVLN：迈向高效的全景视觉-语言导航
PanoVLN: Towards Effective Panoramic Vision-and-Language Navigation
arXiv 2609.34759 · 2026-09-29木星极光区域中质子与电子的沉降：基于Juno/JEDI数据的统计比较
The precipitation of protons and electrons in Jupiter's auroral regions: a statistical comparison based on Juno/JEDI data
arXiv 2609.34758 · 2026-09-29高斯设计下泊松模型中最大似然估计的有限样本行为
Finite sample behavior of the maximum likelihood estimator in the Poisson model under Gaussian design
arXiv 2609.34757 · 2026-09-29从预训练初始化微调的统计优势：对角线性网络
Statistical Benefits of Fine-Tuning from Pretrained Initialization in Diagonal Linear Networks
arXiv 2609.34756 · 2026-09-29结构化气液接触器中界面稳定性的相场模拟：气体扩散电极的设计规则
Phase-field simulation of interfacial stability in structured gas-liquid contactors: design rules for gas diffusion electrodes
arXiv 2609.34755 · 2026-09-29Draft-KV：在语言模型之间学习有用的潜在通信
Draft-KV: Learning Useful Latent Communication Between Language Models
arXiv 2609.34754 · 2026-09-29通过自动自归约的分布式下界
Distributed Lower Bounds via Automatic Self-Reduction
arXiv 2609.34753 · 2026-09-29Multi-SWT-Bench：用于复现测试生成的多语言基准
Multi-SWT-Bench: A Multilingual Benchmark for Reproduction Test Generation
arXiv 2609.34752 · 2026-09-29双曲空间中 $p$-凸超曲面的渐近 Plateau 问题
The Asymptotic Plateau Problem for $p$-Convex Hypersurfaces in Hyperbolic Space
arXiv 2609.34751 · 2026-09-29
