2026-10论文中文摘要
EWAM：统一具身模型中的涌现深度专业化——从语义理解经由视觉预见再到行动
EWAM: Emergent Depth-Wise Specialization in a Unified Embodied Model -- From Semantic Understanding through Visual Foresight to Action
arXiv 2609.39973 · 2026-10-01UBTree：基于一元和二元模型的并行树状草稿生成用于投机解码
UBTree: Parallel Tree Drafting via Unigram and Bigram Models for Speculative Decoding
arXiv 2609.39972 · 2026-10-01当指令检索轨迹：诊断并缓解VLA模型中的泛化失败
When Instructions Retrieve Trajectories: Diagnosing and Mitigating Generalization Failures in VLA Models
arXiv 2609.39971 · 2026-10-01PhasePlan：面向机器人脑模型的有序未来阶段规划
PhasePlan: Ordered Future-Phase Planning for Robot Brain Models
arXiv 2609.39970 · 2026-10-01TACTIC：面向路侧激光雷达攻击的时序与上下文感知大语言模型战术规划
TACTIC: Temporal and Context-Aware LLM Tactical Planning for Roadside LiDAR Attacks
arXiv 2609.39969 · 2026-10-01METANET 模型的输入敏感性与动态标定的鲁棒性
On the Input Sensitivity of METANET Models and the Robustness of Dynamic Calibration
arXiv 2609.39968 · 2026-10-01什么限制了递归推理模型：优化、架构与测试时扩展
What Limits Recursive Reasoning Models: Optimization, Architecture and Test-Time Scaling
arXiv 2609.39967 · 2026-10-01Diptych：音乐制作中参考聆听的限定范围、AI 解释比较
Diptych: Scoped, AI-Interpreted Comparison for Reference Listening in Music Production
arXiv 2609.39963 · 2026-10-01计算流体力学的量子算法
Quantum Algorithms for Computational Fluid Dynamics
arXiv 2609.39962 · 2026-10-01Haah的三维立方码快速热化
Haah's 3D cubic code thermalizes rapidly
arXiv 2609.39959 · 2026-10-01更好的演示文稿还是不同的评审者？评估企业与投资银行中智能体框架的增益
Better Deck or Different Judge? Evaluating Agentic Harness Gains in Corporate and Investment Banking
arXiv 2609.39958 · 2026-10-01学习何时以及如何干预：面向编码智能体的后见之明蒸馏哨兵
Learning When and How to Intervene: A Hindsight-Distilled Sentinel for Coding Agents
arXiv 2609.39957 · 2026-10-01分布式ISAC系统中的资源分配：一种位置敏感的模式选择与任务分配框架
Resource allocation in Distributed ISAC Systems: A Location Sensitive Mode Selection and Task Assignment Framework
arXiv 2609.39956 · 2026-10-01覆盖优先于控制：面向可控逆合成的路线指令接地与引导
Coverage Before Control: Route-Instruction Grounding and Steering for Controllable Retrosynthesis
arXiv 2609.39955 · 2026-10-01双光子关联的共聚焦成像
Confocal imaging from biphoton correlations
arXiv 2609.39954 · 2026-10-01学习在压缩上下文中推理：通过自蒸馏实现OmniLLMs的无真值适配
Learning to Reason with Compressed Context: Ground-Truth-Free Adaptation of OmniLLMs via Self-Distillation
arXiv 2609.39953 · 2026-10-01避免4321及另一个长度为四的模式的对合
Involutions avoiding 4321 and another pattern of length four
arXiv 2609.39952 · 2026-10-01Tate Rabinowitz Floer 同调中的谱数
Spectral numbers in Tate Rabinowitz Floer homology
arXiv 2609.39951 · 2026-10-01基于IMU融合的模块化软体腱驱动连续体机器人磁性原位自三维姿态估计
Magnetic based In-situ Self 3D Pose Estimation for a Modular Soft Tendon-Driven Continuum Robot via IMU-Fusion
arXiv 2609.39950 · 2026-10-01变分量子算法的标度律
Scaling law of variational quantum algorithms
arXiv 2609.39949 · 2026-10-01非线性动态边界反馈下二阶端口-哈密顿系统的输入-状态稳定性
Input-to-state stability of second-order port-Hamiltonian systems under nonlinear dynamic boundary feedback
arXiv 2609.39948 · 2026-10-01铁磁/反铁磁双层结构中的热磁振子输运
Thermal magnon transport in FM/AFM bilayers
arXiv 2609.39947 · 2026-10-01计算群体感应反应网络中的灭绝势垒
Computing Extinction Barriers in a Quorum-Sensing Reaction Network
arXiv 2609.39946 · 2026-10-01边界 Yamabe 问题中紧致性的临界维数，I
On critical dimensions for compactness in the boundary Yamabe problem, I
arXiv 2609.39945 · 2026-10-01大高斯-博内d维黑洞的灰体因子
Greybody factors of large Gauss-Bonnet d-dimensional black holes
arXiv 2609.39944 · 2026-10-01四次平面 Cremona 映射的二次长度
On the quadratic length of plane Cremona maps of degree 4
arXiv 2609.39943 · 2026-10-01关于受控集与微观集的进一步研究
More on dominated and microscopic sets
arXiv 2609.39942 · 2026-10-01速度弥散在Kennicutt-Schmidt关系中的作用
The role of velocity dispersion in the Kennicutt-Schmidt relation
arXiv 2609.39940 · 2026-10-01电磁探针对并发微喷注-流体动力学演化的探测
Electromagnetic probes of concurrent minijet-hydrodynamics evolution
arXiv 2609.39939 · 2026-10-01LEAP：面向长音视频感知的分块证据检索
LEAP: Learned Block-wise Evidence Retrieval for Long Audio-Video Perception
arXiv 2609.39938 · 2026-10-01来自数据的全局认证不变椭球控制
Globally Certified Invariant-Ellipsoid Control from Data
arXiv 2609.39937 · 2026-10-01少数约束多数：多体量子系统的关联增强学习
A Few Constrain Many: Correlation-Enhanced Learning of Many-Body Quantum Systems
arXiv 2609.39935 · 2026-10-01冲突指南：当竞争行为变得可见时，自动研究性能提升
ConflictGuide: AutoResearch Improves When Competing Behaviors Are Made Visible
arXiv 2609.39933 · 2026-10-01军人退伍后的食物不安全问题
Food Insecurity Among Military Veterans
arXiv 2609.39932 · 2026-10-01工程化晶界相称性以实现铪基薄膜中铁电稳定性
Engineering Grain Boundary Commensurability for Ferroelectric Stabilization in Hafnia-Based Films
arXiv 2609.39931 · 2026-10-01大间隙与正整数简并度模谱中的BTZ熵
Large gaps and BTZ entropy in modular spectra with positive integer degeneracies
arXiv 2609.39930 · 2026-10-01RoPE 穷途末路？长上下文失败的理论、诊断与缓解
RoPE at the End of Its Rope? Theory, Diagnosis, and Mitigation of Long-Context Failures
arXiv 2609.39929 · 2026-10-01近场超快无衍射激光纳米级切割玻璃，界面粗糙度准零
Near-field Ultrafast Nondiffractive Laser Nanoscale Cutting of Glass with Quasi-null Interface Roughness
arXiv 2609.39928 · 2026-10-01AdaGEPA：用于反思性提示优化的自适应反馈分配
AdaGEPA: Adaptive Feedback Allocation for Reflective Prompt Optimization
arXiv 2609.39927 · 2026-10-01通过空间算子以任意尺度超分辨未见过的高光谱传感器
Super-Resolving Unseen Hyperspectral Sensors at Any Scale via Spatial Operators
arXiv 2609.39926 · 2026-10-01《Gaia》DR3 非单星星表中年轻星团内的高优先级致密天体候选体
High-Priority Compact-Object Candidates in Young Clusters from the \textit{Gaia} DR3 Non-Single-Star Catalog
arXiv 2609.39925 · 2026-10-01CoVisco：面向统一图像-视频理解的原生码率视觉编码器与原生令牌压缩
CoVisco: Codec-Native Vision Encoder with Native Token Compression for Unified Image-Video Understanding
arXiv 2609.39924 · 2026-10-01通过暗物质衰变提高再电离的光学深度
Raising the Optical Depth to Reionization with Dark Matter Decay
arXiv 2609.39923 · 2026-10-01近对称优化问题上浅层QAOA的容错代价
Fault-tolerant cost of shallow QAOA on near-symmetric optimization problems
arXiv 2609.39922 · 2026-10-01加权幂平均二次型的锐二次上界
Sharp Quadratic Majorants of Weighted Power Means of Quadratic Forms
arXiv 2609.39921 · 2026-10-01MCD：大型视觉-语言模型中多模态上下文学习的大语言模型因果蒸馏
MCD: Causal Distillation of Multimodal In-Context Learning in Large Vision-Language Models
arXiv 2609.39920 · 2026-10-01带常数 $e$ 的随机 Carleson 嵌入定理与 Riesz 变换向量
A stochastic Carleson embedding theorem with constant $e$ and the vector of Riesz transforms
arXiv 2609.39919 · 2026-10-01基于具有植入结构的多项式的可验证量子优势
Verifiable quantum advantage based on polynomials with planted structures
arXiv 2609.39918 · 2026-10-01对数凹性与全幺模多胞体的近似计数
Log-concavity and Approximate Counting for Totally Unimodular Polytopes
arXiv 2609.39917 · 2026-10-01星型网络拓扑上N量子比特稳定子态的蒸馏
Distillation of N-Qubit Stabilizer States on a Star Network Topology
arXiv 2609.39916 · 2026-10-01
