2026-10论文中文摘要
池化表示自编码器用于高效扩散
Pooling Representation Autoencoders for Efficient Diffusion
arXiv 2610.09242 · 2026-10-08对抗图像劫持网页代理：从视觉定位到浏览器执行
Adversarial Images Hijack Web Agents from Visual Grounding to Browser Execution
arXiv 2610.09240 · 2026-10-08LLM 自我改进循环中的赢家诅咒：选择噪声、锁定与接受规则
The Winner's Curse in LLM Self-Improvement Loops: Selection Noise, Lock-in, and Acceptance Rules
arXiv 2610.09239 · 2026-10-08NLSM振幅的BCFW边界项递推关系
Recursion Relation for BCFW Boundary Terms of NLSM Amplitudes
arXiv 2610.09238 · 2026-10-08语言智能体行为的轨迹抽象科学
Trajectory Abstraction for the Science of Language Agent Behavior
arXiv 2610.09237 · 2026-10-08用蒙特卡洛模拟确定pp碰撞喷注区“奇异数增强”的起源
Pinning down the origin of "strangeness enhancement" in the jet-like region of pp collisions with Monte Carlo simulations
arXiv 2610.09236 · 2026-10-08流体混合与不可压缩最优输运
Fluid Mixing and Incompressible Optimal Transport
arXiv 2610.09235 · 2026-10-08自删失模型下信息性缺失的纵向因果推断
Longitudinal causal inference under informative missingness using the self-censoring model
arXiv 2610.09234 · 2026-10-08近似动态时间规整的细粒度难度
Fine-Grained Hardness of Approximating Dynamic Time Warping
arXiv 2610.09233 · 2026-10-08非局域激子光学响应对单层过渡金属硫族化合物面内激子极化激元的影响
Influence of nonlocal excitonic optical response on in-plane exciton polaritons supported by monolayer transition-metal dichalcogenides
arXiv 2610.09232 · 2026-10-08机器学习多群截面可靠不确定性估计（无需重训练）
Reliable Uncertainty Estimation for Machine-Learned Multigroup Cross Sections Without Retraining
arXiv 2610.09231 · 2026-10-08对称性信息下的因果部分识别
Symmetry-Informed Causal Partial Identification
arXiv 2610.09230 · 2026-10-08条件准确率画像：诊断不同部署条件下的LLM裁判
Conditional Accuracy Profiles: Diagnosing LLM Judges across Deployment Conditions
arXiv 2610.09229 · 2026-10-08通过部署协同进化机器人编排器与策略
Co-Evolving Robot Orchestrators and Policies through Deployment
arXiv 2610.09228 · 2026-10-08按漂移量化：文本嵌入器的无标签混合精度训练后量化
Quantize by Drift: Label-Free Mixed-Precision Post-Training Quantization for Text Embedders
arXiv 2610.09227 · 2026-10-08FLoRa：能量约束下从占空比循环LoRa节点进行飞行辅助数据收集
FLoRa: Flight-Assisted Data Collection from Duty Cycling LoRa Nodes under Energy Constraints
arXiv 2610.09226 · 2026-10-08Lyapunov启发的LyRIC激活与GLARE注意力在混沌引导的状态空间建模用于肌电到语音（ETS）合成
Lyapunov-Inspired LyRIC Activation and GLARE Attention in Chaos-Guided State Space Modeling for EMG-To-Speech (ETS) Synthesis
arXiv 2610.09225 · 2026-10-08多物体多目标投掷的快速规划
Fast Planning for Multi-object Multi-target Throwing
arXiv 2610.09224 · 2026-10-08PVSync：用于时序与发音的统一唇形同步专家
PVSync: A Unified Lip-Sync Expert for Timing and Articulation
arXiv 2610.09223 · 2026-10-08扭转SrTiO3薄膜的准晶氧化物界面
Quasicrystalline oxide interfaces by twisting SrTiO3 membranes
arXiv 2610.09222 · 2026-10-08无数据扩散蒸馏的一致性分布匹配
Consistent Distribution Matching for Data-Free Diffusion Distillation
arXiv 2610.09221 · 2026-10-08分布漂移下的遗憾最优线性二次控制
Regret-Optimal Linear-Quadratic Control under Distributional Drift
arXiv 2610.09220 · 2026-10-08CM-DPO：用于LLM规划的约束边际直接偏好优化
CM-DPO: Constraint-Margin Direct Preference Optimization for LLM Planning
arXiv 2610.09219 · 2026-10-08RLDISCOVER：强化学习算法的LLM驱动协同进化
RLDISCOVER: LLM-driven co-evolution of reinforcement learning algorithms
arXiv 2610.09218 · 2026-10-08用于自然场景家庭视频的自闭症谱系筛查的确定性证据层视觉-语言模型
A Deterministic Evidence Layer for Vision-Language Autism Screening from Naturalistic Home Video
arXiv 2610.09217 · 2026-10-08经典软对数定理在$G$领头阶下的大质量粒子轨迹的对数修正
Logarithmic corrections to massive particle trajectories for the classical soft log theorems at leading order in $G$
arXiv 2610.09216 · 2026-10-08AGAR：用于LLM程序进化的强化学习基板
AGAR: a reinforcement learning substrate for LLM program evolution
arXiv 2610.09215 · 2026-10-08无标定鲁棒贝叶斯推断用于离散非归一化模型
Calibration-Free Robust Bayesian Inference for Discrete Unnormalized Models
arXiv 2610.09214 · 2026-10-08关键任务频谱共享中的去中心化多智能体强化学习
Mission-critical spectrum sharing with decentralized Multi-Agent Reinforcement Learning
arXiv 2610.09213 · 2026-10-08CurveTQ：基于曲率加权搜索的无旋转 LLM 权重网格量化
CurveTQ: Rotation-Free Trellis Quantization of LLM Weights via Curvature-Weighted Search
arXiv 2610.09212 · 2026-10-08BanglaBox：面向孟加拉语文本到语音的语音平衡语料库与数据高效基础模型适配，支持零样本声音克隆
BanglaBox: A Phonetically-Balanced Corpus and Data-Efficient Foundation-Model Adaptation for Bangla Text-to-Speech with Zero-Shot Voice Cloning
arXiv 2610.09211 · 2026-10-08三层镍酸盐Pr$_4$Ni$_3$O$_{10}$中的多轨道各向异性与磁激发
Multiorbital Anisotropy and Magnetic Excitations in Trilayer Nickelate Pr$_4$Ni$_3$O$_{10}$
arXiv 2610.09210 · 2026-10-08面向物质使用障碍患者对话生成的多目标对齐小型语言模型框架
Multi-Objective Aligned Small Language Model Framework for SUD Patient Dialogue Generation
arXiv 2610.09209 · 2026-10-08协变量偏移下保形预测的草图校准
Sketched Calibration for Conformal Prediction under Covariate Shift
arXiv 2610.09208 · 2026-10-08间歇观测下基于信念空间规划与规划器条件估计器误差
Belief-Space Planning with Planner-Conditioned Estimator Error under Intermittent Observations
arXiv 2610.09207 · 2026-10-08视觉模型学习的是物理约束还是渲染捷径？一个用于基础物理一致性的反事实基准
Do Vision Models Learn Physical Constraints or Rendering Shortcuts? A Counterfactual Benchmark for Grounded Physical Consistency
arXiv 2610.09205 · 2026-10-08长微通道中活性杆的污染动力学
Contamination dynamics of active rods in long microchannels
arXiv 2610.09204 · 2026-10-08FlexGrad：一种无根式的自适应步长方法
FlexGrad: a root-free approach to adaptive step sizes
arXiv 2610.09203 · 2026-10-08少比特，一法则：迈向 W2A4KV2
Few Bits, One Law: Toward W2A4KV2
arXiv 2610.09202 · 2026-10-08Cabling公式与彩色纽结Floer同调
Cabling formula and colored knot Floer homology
arXiv 2610.09201 · 2026-10-08StyleFields：用于从粗到细三维形状重建与编辑的多尺度AdaIN调制隐式SDF
StyleFields: Multi-Scale AdaIN-Modulated Implicit SDFs for Coarse-to-Fine 3D Shape Reconstruction and Editing
arXiv 2610.09200 · 2026-10-08对GW190521性质的完整理解
Toward a complete understanding of the properties of GW190521
arXiv 2610.09199 · 2026-10-08Rust 异步运行时的模块化响应性验证
Modular Responsiveness Verification of Rust Async Runtimes
arXiv 2610.09198 · 2026-10-08FreeEvolve：学习超越固定循环的进化
FreeEvolve: Learning to Evolve Beyond Fixed Loops
arXiv 2610.09197 · 2026-10-08线性递归特征机的精确动力学与有限样本轨迹恢复
Exact Dynamics and Finite-Sample Trajectory Recovery of Linear Recursive Feature Machines
arXiv 2610.09196 · 2026-10-08StableGrasp：从单张图像重建物理稳定的人手抓取
StableGrasp: Reconstructing Physically Stable Human Hand Grasps from Single Images
arXiv 2610.09195 · 2026-10-08患者、地点、先验（P$^3$）：医学世界模型中的个性化意味着什么？
Patient, Place, Prior (P$^3$): What Counts as Personalization in Medical World Models?
arXiv 2610.09194 · 2026-10-08记账、组合，还是不可达的金矿？对照冻结的最后写入解析器解读 MemoryAgentBench 的冲突解决分数
Bookkeeping, Composition, or Unreachable Gold? Reading MemoryAgentBench's Conflict-Resolution Scores Against a Frozen Last-Write Resolver
arXiv 2610.09193 · 2026-10-08多层热反射测量中快速不确定性传播的物理推导自然坐标
Physics-Derived Natural Coordinates for Fast Uncertainty Propagation in Multilayer Thermoreflectance
arXiv 2610.09192 · 2026-10-08基于R-CNN的棋盘局面识别
R-CNN-Based Chess Position Recognition
arXiv 2610.09191 · 2026-10-08
