2026-10论文中文摘要
从视频片段到创作轨迹：面向AI原生视频创作的Sora100K
From Video Clips to Creation Trajectory: Sora100K for AI-Native Video Creation
arXiv 2610.11770 · 2026-10-09Lawson Machine 26上1 keV电子温度的汤姆孙散射测量
Thomson Scattering Measurement of 1 keV Electron Temperature on Lawson Machine 26
arXiv 2610.11769 · 2026-10-09窄而深：作为工业设备系统大语言模型智能体知识的本体塔
Narrow and Deep: An Ontology Tower as the Knowledge of an LLM Agent for an Industrial Equipment System
arXiv 2610.11768 · 2026-10-09基于GW-贝特-萨尔佩特框架的数千原子体系光吸收光谱计算
Optical absorption spectra from the $GW$-Bethe-Salpeter framework for thousands of atoms
arXiv 2610.11767 · 2026-10-09相同结果，不同证据：大语言模型安全评估中的意图恢复
Same Outcome, Different Evidence: Intent Recovery in LLM Safety Evaluation
arXiv 2610.11766 · 2026-10-09思维惯性：被要求停止思考时大语言模型仍会继续思考
Thinking Inertia: LLMs Keep Thinking When Told Not To
arXiv 2610.11765 · 2026-10-09UltraLight Luma：一种用于农业机器人作物行分割的新型边缘可部署感知网络
UltraLight Luma: A Novel Edge-Deployable Perception Network for Crop-Row Segmentation in Agricultural Robotics
arXiv 2610.11764 · 2026-10-09金属丰度不会重新排序类太阳恒星中谱线对米粒组织的线间灵敏度
Metallicity does not reorder line-to-line sensitivity to granulation in solar-like stars
arXiv 2610.11763 · 2026-10-09利用TESS食双星的本征多重性约束探测大质量恒星的形成
Probing Massive Star Formation with Intrinsic Multiplicity Constraints from TESS Eclipsing Binaries
arXiv 2610.11762 · 2026-10-09RAGenome：将基于检索的基因组语言模型扩展至长上下文
RAGenome: Scaling Retrieval-Based Genomic Language Models to Long Contexts
arXiv 2610.11761 · 2026-10-09自适应极小极大多元L¹反卷积：噪声滤波下的调和平均速率与1-瓦瑟斯坦等价性
Adaptive minimax multivariate \(L^1\)-deconvolution: harmonic-mean rates under noise filtering, and a \(1\)-Wasserstein equivalence
arXiv 2610.11760 · 2026-10-09超越QAOA：人工智能与量子计算在自适应组合优化中的综述
Beyond QAOA: A Review of AI and Quantum Computing for Adaptive Combinatorial Optimization
arXiv 2610.11759 · 2026-10-09AFDM信道参数估计：利用其移位特性与输入输出关系
AFDM Channel Parameter Estimation: Exploiting its Shift Properties and Input-Output Relationship
arXiv 2610.11758 · 2026-10-09超图扩张的Turán密度
Turán densities of hypergraph augmentations
arXiv 2610.11757 · 2026-10-09记忆强制：用于流式视频生成的可关注中 horizon 历史
Memory Forcing: Attendable Mid-Horizon History for Streaming Video Generation
arXiv 2610.11756 · 2026-10-09SR-TTA：面向抗干扰呼吸感知的空间冗余测试时自适应
SR-TTA: Spatial-Redundancy Test-Time Adaptation for Interference-Robust Respiration Sensing
arXiv 2610.11755 · 2026-10-09仅输出内容无法验证的问题：研究智能体的研究契约
What Output-Only Review Cannot Verify: Study Contracts for Research Agents
arXiv 2610.11754 · 2026-10-09直觉主义策梅洛-弗兰克尔集合论的前提独立性
Independence of premise for intuitionistic Zermelo-Fraenkel set theory
arXiv 2610.11753 · 2026-10-092DGS-Planner：基于光栅化的2D高斯溅射地图路径规划器
2DGS-Planner: Rasterization-based Path Planning in 2D Gaussian Splatting Map
arXiv 2610.11752 · 2026-10-09Dino Forcing Flow Models：不要去噪你可以预测的内容
Dino Forcing Flow Models: Do not denoise what you can predict
arXiv 2610.11751 · 2026-10-09草稿树丢失目标质量的原因：出口引导的推测解码
Where Draft Trees Lose Target Mass: Exit-Guided Speculative Decoding
arXiv 2610.11750 · 2026-10-09相干谐波求和脉冲星搜索代码
A Coherent Harmonic Summing Pulsar Search Code
arXiv 2610.11749 · 2026-10-09DEX：面向高能效MSDF神经网络推理的数字级早期退出机制
DEX: Digit-Level Early Exit for Energy-Efficient MSDF Neural Network Inference
arXiv 2610.11748 · 2026-10-09树张量网络的鲁棒二阶时间积分方法
Robust second-order time integration of tree tensor networks
arXiv 2610.11747 · 2026-10-09面向实时视频超分辨率的流感知扩散模型：基于跨步注意力机制
Streaming-Aware Diffusion for Real-Time Video Super-Resolution via Cross-Step Attention
arXiv 2610.11746 · 2026-10-09核子张量多极矩的味分解
Flavour decomposition of the nucleon tensor multipole moments
arXiv 2610.11745 · 2026-10-09球与盒：叠加态下的两种计算几何结构
The Ball and the Box: Two Geometries of Computation in Superposition
arXiv 2610.11744 · 2026-10-09TraceRelay：基于滚动轨迹的注意力对齐循环架构
TraceRelay: Attention-Aligned Recurrence over Rolling Traces
arXiv 2610.11743 · 2026-10-09从原始电子结构进行端到端学习能否解释磁各向异性？
Can end-to-end learning from raw electronic structure explain magnetic anisotropy?
arXiv 2610.11742 · 2026-10-09从超立方体的稀疏观测中高效恢复潜在坐标结构
Efficient Recovery of Latent Coordinate Structure from Sparse Observations of the Hypercube
arXiv 2610.11741 · 2026-10-09形态神经网络的关联训练
Correlational Training of Morphological Neural Networks
arXiv 2610.11740 · 2026-10-096G网络中的分布式受限资源管理：一种可扩展的混合模型学习框架
Distributed Constrained Resource Management in 6G Networks: A Scalable Hybrid Model-Learning Framework
arXiv 2610.11739 · 2026-10-09基于广义拉普拉斯主动子空间的可扩展AI不确定性量化
Scalable AI Uncertainty Quantification via Generalized Laplace Active Subspaces
arXiv 2610.11738 · 2026-10-09揭示并修正贝叶斯物理信息神经网络中的碰撞器偏差
Uncovering and Fixing Collider Bias in Bayesian PINNs
arXiv 2610.11737 · 2026-10-09面向视频生成提示增强器的统一评估
Towards Unified Evaluation of Prompt Enhancers for Video Generation
arXiv 2610.11736 · 2026-10-09Φsat-2星载海洋异常检测：从基于仿真的开发到在轨演示
Onboard Marine Anomaly Detection on $Φ$sat-2: From Simulation-Based Development to In-Orbit Demonstration
arXiv 2610.11735 · 2026-10-09Timer-M1：一种通过学习基元实现的多元时间序列基础模型
Timer-M1: A Multivariate Time Series Foundation Model via Learning Primitives
arXiv 2610.11734 · 2026-10-09弹性常数的最小非耦合应力-应变计算通用算法
A General Algorithm for Minimal Uncoupled Stress--Strain Calculations of Elastic Constant
arXiv 2610.11733 · 2026-10-09MemTrial：学习何时信任LLM组合智能体中的记忆
MemTrial: Learning When to Trust Memory in LLM Portfolio Agents
arXiv 2610.11732 · 2026-10-09Fe₃GaTe₂中压力增强层间交换耦合揭示的层间铁磁有序
Interlayer ferromagnetic order revealed through pressure-enhanced interlayer exchange coupling in Fe$_3$GaTe$_2$
arXiv 2610.11731 · 2026-10-09谱权重衰减：诱导神经网络权重的低秩结构
Spectral Weight Decay: Inducing Low-Rank Structure in Neural Network Weights
arXiv 2610.11730 · 2026-10-09利用X射线激光对真空中的孤立膜蛋白复合物进行三维成像
Three-dimensional imaging of isolated membrane-protein complexes in vacuo with an X-ray laser
arXiv 2610.11729 · 2026-10-09拟Coxeter元的非交叉组合学
Noncrossing combinatorics of quasi-Coxeter elements
arXiv 2610.11728 · 2026-10-09MAP4CS：用于高效代码检索器微调的多维数据剪枝框架
MAP4CS: A Multi-dimensional Data Pruning Framework for Efficient Code Retriever Fine-tuning
arXiv 2610.11727 · 2026-10-09正曲率真空静态空间的刚性
Rigidity of positively curved vacuum Static spaces
arXiv 2610.11726 · 2026-10-09实现规范增长引擎：防止规范-代码分歧并通过智能体扩展规范
Implementing the Spec Growth Engine: Preventing Spec-Code Divergence, and Growing the Spec with Agents
arXiv 2610.11725 · 2026-10-09解决多模态歧义下性别化经济梗推理中的过度承诺问题
Addressing Overcommitment in the Reasoning of Gendered Economic Memes under Multimodal Ambiguity
arXiv 2610.11724 · 2026-10-09MultiWorldBench：独立控制的视图是否描述同一个共享世界？
MultiWorldBench: Do Independently Controlled Views Describe One Shared World?
arXiv 2610.11723 · 2026-10-09从单智能体到多智能体集成：面向可组合多智能体人-物体交互的分层框架
From Solo to Ensemble: A Hierarchical Framework for Composable Multi-Agent Human-Object Interaction
arXiv 2610.11722 · 2026-10-09SubDGuide：一种受建模者启发的用于网格到细分曲面（SubD）重建的智能体工作流
SubDGuide: A Modeler-Inspired Agentic Workflow for Mesh-to-SubD Reconstruction
arXiv 2610.11721 · 2026-10-09
