arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

Imperial College London(帝国理工学院)

共收录 180
2607.02799 2026-07-07 cs.CV 新提交

Conversational Human Audio-visual Talking Dialogue Generation

对话式人类视听对话生成

Junhao Song, Lluis Guasch, Xilin He, Zhongyu Yang, Yingfang Yuan, Weicheng Xie, Linlin Shen, Haijun Lin, Shizhe Liu, Wei Pang, Siyang Song

机构 * Department of Computing, Imperial College London(伦敦帝国理工学院计算系) Department of Earth Science & Engineering, Imperial College London(伦敦帝国理工学院地球科学与工程系) Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学) Department of Computer Science, Heriot-Watt University(赫瑞瓦特大学计算机科学系) School of Computer Science, Northumbria University(诺森比亚大学计算机科学学院) College of Computer Science & Software Engineering, Shenzhen University(深圳大学计算机科学与软件学院) School of Artificial Intelligence, Shenzhen University(深圳大学人工智能学院) Guangdong Provincial Key Laboratory of Intelligent Information Processing, Shenzhen University(深圳大学广东省智能信息处理重点实验室) School of Engineering and Design, Hunan Normal University(湖南师范大学工程与设计学院) Department of Computer Science, University of Oxford(牛津大学计算机科学系) Department of Computer Science, University of Exeter(埃克塞特大学计算机科学系)

AI总结 提出CHAT框架,统一大语言模型和说话人脸模型,通过交互音频和面部行为细化模块,从单一文本提示生成多样、配对且相互响应的语音-面部对话片段,优于现有方法,合成数据集可作预训练数据。

Comments Accepted to ECCV 2026 as a main paper

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.02571 2026-07-07 cs.CV 新提交

Dual-Adaptive SAM3: Hierarchical Routing over Low-Rank Expert Layers for Parameter-Efficient Medical Image Segmentation

双自适应SAM3:基于低秩专家层的分层路由用于参数高效的医学图像分割

Ying Chen, Jinyue Li, Kun Wang, Qiankun Li, Yang Liu

机构 * Shenzhen Research Institute, The Chinese University of Hong Kong(香港中文大学深圳研究院) University of Science and Technology of China(中国科学技术大学) Nanyang Technological University(南洋理工大学) Imperial Global Singapore (IGS), Imperial College London(伦敦帝国理工学院新加坡帝国全球(IGS))

AI总结 提出双自适应SAM3框架,通过任务感知的动态专家路由器和参数感知的分解参数化专家设计,兼顾分割精度与参数效率,在医学图像分割上有高表现。

Comments Accepted by MICCAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.02560 2026-07-07 cs.CV 新提交

Inpainting U-Net for seamless pedestrian-level wind prediction across urban morphologies

用于跨城市形态的无缝行人级风速预测的修复U-Net

Jingzi Huang, Claire E. Heaney, Tao Li, Xinzhe Li, Graham O. Hughes, Maarten van Reeuwijk

机构 * Department of Civil and Environmental Engineering, Imperial College London(伦敦帝国理工学院土木与环境工程系) Department of Earth Science and Engineering, Applied Modelling and Computation Group, Imperial College London(伦敦帝国理工学院地球科学与工程系应用建模与计算组)

AI总结 研究针对行人级风速预测,开发两阶段U-Net框架。先由基线U-Net逐块预测,再用基于修复的U-Net改进,减少块边界不连续性,为跨城市形态的高分辨率行人级风速预测提供了高效灵活的替代模型。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.03168 2026-07-07 math.OC cs.LG 新提交

Entropy Regularization Improves Policy Robustness in Continuous-Time Reinforcement Learning

熵正则化提高连续时间强化学习中的策略鲁棒性

Jialun Cao, Fernando Acero, David Šiška, Yufei Zhang

机构 * University of Edinburgh(爱丁堡大学) J.P. Morgan AI Research(摩根大通人工智能研究) Imperial College London(伦敦帝国理工学院)

AI总结 研究连续时间强化学习中熵正则化的策略鲁棒性,建立相关理论保证,分析鲁棒集特性,实验验证其有效性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.03278 2026-07-07 quant-ph cs.CC cs.LG 新提交

Complexity of Normalized Persistence Problems for Topological Data Analysis and Local Hamiltonians

拓扑数据分析和局部哈密顿量的归一化持久性问题的复杂性

Dominic Lowe, M. S. Kim, Roberto Bondesan, Ryu Hayakawa

机构 * Blackett Laboratory, Imperial College London(帝国理工学院伦敦分校黑克特实验室) Department of Computing, Imperial College London(帝国理工学院伦敦分校计算系) Yukawa Institute for Theoretical Physics & The Hakubi Center, Kyoto University(京都大学山梨研究所及 Hakubi 中心)

AI总结 研究归一化持久性问题,证明其变体是$\mathsf{DQC}_1$-难且含于$\mathsf{BQP}$,揭示与局部哈密顿量低能子空间谱量估计复杂性的联系,还介绍$\mathsf{SDQC}_1$刻画精确核归一化问题的硬度。

Comments 61 pages, 4 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.02148 2026-07-03 cs.CV 新提交

SAMoR: Motion Modelling for Articulated Objects of Any Skeleton and Topology

SAMoR: 任意骨架和拓扑结构的铰接物体运动建模

Yuhao Zhang, Gerard Pons-Moll, Tolga Birdal

机构 * Imperial College London(伦敦帝国学院) University of Tübingen(图宾根大学) Tübingen AI Center(图宾根人工智能中心)

AI总结 提出SAMoR,一种跨拓扑的运动表示方法,将运动片段编码为固定数量的部件令牌,实现任意骨架的精确重建和跨拓扑运动迁移。

Comments 20 pages, 5 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.01449 2026-07-03 cs.LG cs.NA math.NA 新提交

Geometry-Aware R-Structured Kolmogorov-Arnold Networks

几何感知的R结构Kolmogorov-Arnold网络

Sergei Kucherenko, Nilay Shah

机构 * Imperial College London(帝国理工学院)

AI总结 提出GRS-KAN混合架构,将R函数集成到KAN中,通过可微逻辑运算编码几何约束,在回归任务中相比标准KAN测试RMSE降低高达67%,并提升可解释性。

Comments 27 pages, 13 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.02363 2026-07-03 quant-ph cs.AI cs.ET cs.LG cs.NE 新提交

Stable Self-Modulating Quantum Fast-Weight Programmers with Bounded Memory Gates

带界内存门的稳定自调制量子快速权重编程器

Kuo-Chung Peng, Jiun-Cheng Jiang, Chun-Hua Lin, Yifeng Peng, Junghoon Justin Park, Huan-Hsin Tseng, Hsin-Yi Lin, Kuan-Cheng Chen, Chen-Yu Liu, Shinjae Yoo, Samuel Yen-Chi Chen

机构 * National Taiwan University(国立台湾大学) Stevens Institute of Technology(史蒂文斯理工学院) Seoul National University(首尔国立大学) Brookhaven National Laboratory(布鲁克海文国家实验室) Imperial College London(伦敦帝国理工学院) Wells Fargo(沃尔士·法戈)

AI总结 提出带界旧态调制规则的自调制量子快速权重编程器,消除长序列发散并提升鲁棒性,在量子动力学预测和电信活动预测任务上验证了其有效性。

Comments 16 pages, 8 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.01147 2026-07-02 cs.CV 新提交

EquiSteer: Cross-Attention Steering Towards a Fairer Text-Guided Image Generation

EquiSteer: 面向更公平的文本引导图像生成的交叉注意力引导

Tatiana Gaintseva, Akshit Achara, Gregory Slabaugh, Jiankang Deng, Ismail Elezi

机构 * Queen Mary University of London(伦敦玛丽女王大学) Huawei Noah’s Ark(华为诺亚方舟实验室) King’s College London(伦敦国王学院) Imperial College London(帝国理工学院)

AI总结 提出EquiSteer,一种无需训练的方法,通过在推理时引导交叉注意力激活来减少文本到图像扩散模型中的性别偏见,平均减少高达87%的性别差异。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.01082 2026-07-02 cs.LG 新提交

When Context Compensates for Sparse Event History: AlphaEarth for Spatio-Temporal Point-Process Forecasting

当上下文弥补稀疏事件历史:AlphaEarth用于时空点过程预测

Yahya Aalaila, Mouad Elhamdi, Gerrit Großmann, Daniel Jenson, Elizaveta Semenova, Sebastian Vollmer

机构 * German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心) Université Mohammed VI Polytechnique(穆罕默德六世理工大学) University of Oxford(牛津大学) Imperial College London(帝国理工学院) Rhineland-Palatinate Technical University of Kaiserslautern-Landau (RPTU)(莱茵兰-普法尔茨凯泽斯劳滕-兰道工业大学)

AI总结 研究利用AlphaEarth嵌入作为空间上下文,在事件历史稀疏时提升时空点过程模型的跨区域预测性能,实验表明在短历史下预测误差降低2-6倍。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.00060 2026-07-02 cs.CV 新提交

Synergistic Perception-Reasoning Governance: Grounding Medical MLLMs with Verifiable Anatomical Evidence

协同感知-推理治理:用可验证解剖证据为医学多模态大语言模型提供基础

Rui Hao, Qiankun Li, Junyuan Mao, Linghao Meng, Dirui Xie, Dayu Tan, Zhigang Zeng

机构 * Huazhong University of Science and Technology(华中科技大学) Imperial Global Singapore, Imperial College London(帝国理工学院新加坡全球中心) Nanyang Technological University(南洋理工大学) National University of Singapore(新加坡国立大学) Anhui University(安徽大学)

AI总结 提出一种无需训练的证据注入框架,通过ROI引导的视觉激活调制和解剖坐标语义标记,协同校准视觉感知与文本推理,动态路由任务特定干预,有效减少医学MLLM的幻觉。

Comments Accepted by MICCAI 2026 (Early Accept, Top 9%)

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.00058 2026-07-02 cs.CV 新提交

Joint Medical Image Enhancement and Segmentation with Diffusion-based Symbiotic Information Interaction

基于扩散的共生信息交互的联合医学图像增强与分割

Ying Chen, Jinyue Li, Qiankun Li

机构 * Shenzhen Research Institute, The Chinese University of Hong Kong(香港中文大学深圳研究院) University of Science and Technology of China(中国科学技术大学) Imperial Global Singapore, Imperial College London(伦敦帝国学院新加坡分校)

AI总结 提出DiSIINet,利用扩散模型和共生信息交互模块,在统一框架中实现图像增强与分割的相互促进,在多模态医学图像上取得显著性能提升。

Comments Accepted by IJCAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.31729 2026-07-01 eess.AS cs.LG 新提交

Is Natural Always Appropriate? Investigating Naturalness and Appropriateness Across Different Domains for TTS Evaluation

自然是否总是合适的?探讨TTS评估中不同领域的自然性与合适性

Dominika Woszczyk, Andreas Triantafyllopoulos, Jura Miniota, Éva Székely, Bjoern Schuller

机构 * Iconic, United Kingdom(Iconic,英国) Technische Universität München, Germany(慕尼黑工业大学,德国) KTH Royal Institute of Technology, Sweden(瑞典皇家理工学院,瑞典) Imperial College London, United Kingdom(伦敦帝国理工学院,英国)

AI总结 研究五个SOTA TTS系统在五个领域的合适性与自然性,发现合适性随领域变化且独立于自然性,表明TTS性能未解决,需上下文感知评估。

Comments Accepted at Interspeech 26'

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.31521 2026-07-01 eess.IV cs.CV eess.SP 新提交

Distortion-Corrected Diffusion MRI Using Rotated-View EPI and Joint Field-Map/Image Estimation with Gaussian Primitives

使用旋转视图EPI和高斯原型的联合场图/图像估计的畸变校正扩散MRI

Wenqi Huang, Zhitao Li, Nan Wang, Yimeng Lin, Mengze Gao, Yurui Qian, Sevgi Gokce Kafali, Xiaozhi Cao, Kawin Setsompop, Daniel Rueckert, Congyu Liao

机构 * Chair for AI in Healthcare and Medicine, Technical University of Munich (TUM) and TUM University Hospital, and the School of Computation, Information and Technology, TUM(慕尼黑工业大学(TUM)医疗与医学人工智能教席、TUM大学医院、计算信息与技术学院) Neuroimaging Technology Research Center, Department of Radiology and Biomedical Imaging, University of California, San Francisco(加州大学旧金山分校放射与生物医学影像系神经影像技术研究中心) Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences(中国科学院精密测量科学与技术创新研究院) Department of Radiology, Stanford University(斯坦福大学放射学系) Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心(MCML)) Department of Computing, Imperial College London(伦敦帝国理工学院计算系)

AI总结 提出一种物理信息框架,直接从k空间联合估计B0场和无畸变图像,避免中间并行成像重建,通过高斯原型的连续参数化表示图像和场,支持旋转视图EPI,在高b值和高加速下显著改善畸变校正质量。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.31635 2026-07-01 eess.SY cs.AI cs.MA cs.SY 新提交

A Tutorial on Autonomous Fault-Tolerant Control Using Knowledge-Grounded LLM Agents

基于知识接地LLM代理的自主容错控制教程

Javal Vyas, Milapji Singh Gill, Artan Markaj, Felix Gehlhoff, Mehmet Mercangöz

机构 * Autonomous Industrial Systems Laboratory, Imperial College London(帝国理工学院自主工业系统实验室) Institute of Automation Technology, Helmut Schmidt University(赫尔穆特·施密特大学自动化技术研究所)

AI总结 提出利用LLM代理作为约束监督规划器,通过外部验证器确保安全,辅助过程工厂故障恢复决策,并提供了可执行的Python环境。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.31614 2026-07-01 eess.SY cs.AI cs.SY 新提交

Automating Cause-Effect Specification with Knowledge Graphs and Large Language Models

利用知识图谱和大语言模型自动化因果规范生成

Javal Vyas, Milapji Singh Gill, Mehmet Mercangöz

机构 * Autonomous Industrial Systems Lab, Imperial College London(帝国理工学院自主工业系统实验室)

AI总结 提出一种语义AI框架,结合知识图谱与约束大语言模型,自动生成因果逻辑和操作安全叙述,减少手动工作。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.28179 2026-06-29 cs.LG cs.AI 新提交

CPAgents: Agentic Composite Phenotype Generation for Cardiac Disease Association

CPAgents: 用于心脏疾病关联的智能复合表型生成

Zuoou Li, Wenlong Zhao, Kelly Yu, Weitong Zhang, Paul M. Matthews, Wenjia Bai, Bernhard Kainz, Mengyun Qiao

机构 * Department of Mechanical Engineering, University College London(伦敦大学学院机械工程系) CSIG Group, Tencent(腾讯CSIG组) Department of Computing, Imperial College London(帝国理工学院计算系) Department of Brain Sciences, Imperial College London(帝国理工学院脑科学系) Data Science Institute, Imperial College London(帝国理工学院数据科学研究所) FAU Erlangen–Nürnberg(埃尔朗根-纽伦堡大学) UK Dementia Research Institute, Imperial College London(英国痴呆症研究所帝国理工学院) Rosalind Franklin Institute(罗莎琳德·富兰克林研究所)

AI总结 提出CPAgents框架,通过多智能体协作自动构建可解释的复合表型(如多项式、比值和交互形式),在群体规模心脏影像队列中显著提升疾病判别性能。

Comments Accepted to MICCAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.28011 2026-06-29 eess.SY cs.LG cs.SY 新提交

From Detection to Action: Using LLM Agents for Fault-Tolerant Control

从检测到行动:使用LLM智能体进行容错控制

Javal Vyas, Milapji Singh Gill, Artan Markaj, Felix Gehlhoff, Mehmet Mercangöz

机构 * Imperial College London(帝国理工学院伦敦校区) Helmut Schmidt University(海德堡-施密特大学)

AI总结 提出基于大语言模型智能体的主动容错控制框架,通过多智能体协作、数字孪生和知识图谱增强检索,生成并验证最小风险恢复路径,在离散和连续过程控制中实现从故障检测到有效纠正行动的闭环。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.25877 2026-06-25 cs.RO 新提交

TacVerse: A Multi-Sensor Dataset and Benchmark for Cross-Sensor Vision-Based Tactile Perception

TacVerse:面向跨传感器视觉触觉感知的多传感器数据集与基准

Lan Wei, Gurmeher Khurana, Sirine Bhouri, Wenhao Hong, Zeyuan Xin, Qingzheng Cong, Wen Fan, Yanzheng Xiang, Dandan Zhang

机构 * Imperial College London(帝国理工学院) Queen Mary University of London(伦敦玛丽女王大学)

AI总结 提出TacVerse多传感器数据集与基准,包含7种视觉触觉传感器的106800张图像,支持形状分类、光栅分类和力回归任务,实验表明跨传感器直接迁移性能下降,而少样本适应可提升力回归性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.23838 2026-06-25 cs.LG astro-ph.IM physics.comp-ph physics.data-an stat.ML 新提交

The Degeneracy Distillery

退化蒸馏器

T. Lucas Makinen, Deaglan J. Bartlett, Niall Jeffrey, Benjamin D. Wandelt

机构 * Department of Applied Mathematics and Theoretical Physics, University of Cambridge(剑桥大学应用数学与理论物理系) Imperial Centre for Inference and Cosmology (ICIC), Imperial College London(伦敦帝国理工学院帝国推理与宇宙学中心) Astrophysics, University of Oxford(牛津大学天体物理学系) CNRS & Sorbonne Université, Institut d’Astrophysique de Paris (IAP)(法国国家科学研究中心与索邦大学巴黎天体物理研究所) Department of Physics and Astronomy, University College London(伦敦大学学院物理与天文学系) Department of Physics & King’s Institute for Artificial Intelligence, King’s College London(伦敦国王学院物理系与国王人工智能研究所) Department of Physics and Astronomy, Johns Hopkins University(约翰霍普金斯大学物理与天文学系) Department of Applied Mathematics and Statistics, Johns Hopkins University(约翰霍普金斯大学应用数学与统计学系)

AI总结 提出退化蒸馏器方法,通过估计和展平Fisher信息矩阵,自动符号化检测并解决物理模型中的退化参数组合,降低神经后验估计所需的模拟预算。

Comments 30 pages, 10 figures. Supporting code found at https://github.com/tlmakinen/degeneracy_distillery

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.24759 2026-06-24 cs.CV cs.AI 新提交

UniDrive: A Unified Vision-Language and Grounding Framework for Interpretable Risk Understanding in Autonomous Driving

UniDrive: 面向自动驾驶可解释风险理解的统一视觉-语言与定位框架

Xiaowei Gao, Pengxiang Li, Yitai Cheng, Ruihan Xu, James Haworth, Stephen Law, Yun Ye

机构 * organization= Department of Earth Science \& Engineering, Imperial College London , city= London , postcode= SW7 2AZ , country= United Kingdom organization= SpaceTimeLab, Department of Civil, Environmental Geomatic Engineering, University College London , city= London , postcode= WC1E 6BT , country= United Kingdom organization= Department of Computing, The Hong Kong Polytechnic University , city= Hong Kong , country= China organization= Trinity College, University of Oxford , city= Oxford , postcode= OX1 3BH , country= United Kingdom organization= Department of Geography, University College London , city= London , postcode= WC1E 6BT , country= United Kingdom organization= Centre for Global Infrastructure Resilience, The Bartlett School of Sustainable Construction, University College London , city= London , postcode= WC1E 7HB , country= United Kingdom

AI总结 提出UniDrive框架,通过融合时序推理与高分辨率感知分支,联合生成风险描述和边界框定位,在DRAMA-Reasoning基准上超越现有方法,提升小目标定位和可解释性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.24457 2026-06-24 cs.CV 新提交

Lite Any Stereo V2: Faster and Stronger Efficient Zero-Shot Stereo Matching

Lite Any Stereo V2:更快更强的零样本立体匹配

Junpeng Jing, Ronglai Zuo, Zhelun Shen, Shangchen Zhou, Rolandos Alexandros Potamias, Stefanos Zafeiriou, Krystian Mikolajczyk, Jiankang Deng

机构 * Imperial College London(帝国理工学院)

AI总结 提出Lite Any Stereo V2超快模型系列,通过2D代价聚合框架和三阶段训练策略(合成监督、自蒸馏、真实知识蒸馏)实现高效零样本立体匹配,在保持低延迟的同时达到最先进精度。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.24367 2026-06-24 cs.SD stat.AP 新提交

Statistical validation and full-sphere extension of a Bayesian model for human static sound localisation

人类静态声源定位贝叶斯模型的统计验证与全空间扩展

Roberto Barumerli, Fabian Brinkmann, Emanuele Zanoni, Anton Hoyer, Lorenzo Picinali, Michele Geronazzo

机构 * Dyson School of Design Engineering, Imperial College London(帝国理工学院戴森设计工程学院) Audio Communication Group, Technische Universität Berlin(柏林工业大学音频通信组) Department of Industrial Systems Technology and Management, University of Padova(帕多瓦大学工业系统技术与管理系)

AI总结 提出贝叶斯声源定位模型的显式似然函数,通过参数恢复和行为数据拟合验证其可靠性,并比较四种HRTF模板插值方法,发现全空间覆盖和高频保真度是关键。

Comments 16 pages, 6 figures, 3 supplementary figures; submitted to Acta Acustica (special issue on Spatial and Binaural Hearing: From Neural Processes to Applications)

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.23742 2026-06-24 cs.LG cs.AI cs.AR 新提交

Low-power analogue neural networks with trainable nonlinear connections for continuous control

具有可训练非线性连接的低功耗模拟神经网络用于连续控制

Ian T. Vidamour, Fernando Aguirre, Thomas J. Hayward, Matthew O. A. Ellis, Charles Swindells, Alexander McDonnell, Martin Trefzer, Finley Robins, Luca Manneschi, Susan Stepney, Tony Kenyon, Oliver J. Sutton, Jack C. Gartside, Ivan Y. Tyukin, Adnan Mehonic, Eleni Vasilaki

机构 * School of Computer Science, University of Sheffield(谢菲尔德大学计算机科学学院) Intrinsic Semiconductor Technologies(Intrinsic Semiconductor Technologies公司) School of Chemical, Biological, and Materials Science Engineering, University of Sheffield(谢菲尔德大学化学、生物与材料科学工程学院) School of Physics, Engineering, and Technology, University of York(约克大学物理、工程与技术学院) Department of Computer Science, University of York(约克大学计算机科学系) Department of Electronic & Electrical Engineering, University College London(伦敦大学学院电子与电气工程系) King’s College London(伦敦国王学院) Blackett Laboratory, Imperial College London(帝国理工学院布莱克特实验室)

AI总结 受Kolmogorov-Arnold网络启发,在连接上放置可训练非线性函数,使每个物理连接成为可学习计算单元,通过现场可编程模拟阵列实现带通滤波器,在连续控制等任务上以更少节点和连接达到高效,预计CMOS实现功耗约30微瓦。

Comments Preprint. Further verification of all simulations is ongoing. Any resulting corrections will be incorporated in a revised version

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.23827 2026-06-24 math.OC cs.LG cs.NA math.NA 新提交

Hessian-augmented Supervised Learning for Hamilton-Jacobi-Bellman PDEs

Hessian增强的Hamilton-Jacobi-Bellman偏微分方程监督学习

Matías Gómez-Aedo, Behzad Azmi, Yuyang Huang, Dante Kalise, Karl Kunisch

机构 * Department of Mathematics, Imperial College London, South Kensington Campus(帝国理工学院伦敦数学系,南肯辛顿校区) Department of Mathematics and Statistics, University of Konstanz(康斯坦茨大学数学与统计学系) RICAM and Institute of Mathematics and Scientific Computing, University of Graz(格拉茨大学RICAM与数学与计算科学研究所)

AI总结 提出一种数据驱动方法,利用最优控制问题中值函数的梯度与Hessian信息增强加权最小二乘回归,显著降低样本复杂度并提高近似精度,在高维问题中采用部分Hessian策略控制成本。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.22601 2026-06-23 stat.ML cs.LG stat.AP stat.CO 新提交

Scalable Bayesian Additive Models for Stellar Flare Detection via Amortized Gaussian Process Inference and Hidden Markov Models

可扩展贝叶斯加性模型:通过摊销高斯过程推理和隐马尔可夫模型进行恒星耀斑检测

Rodrigo Herrera, Vianey Leos-Barajas, Gwendolyn Eadie, Elizaveta Semenova, James Davenport

机构 * Department of Statistical Sciences, University of Toronto(多伦多大学统计科学系) Data Sciences Institute, University of Toronto(多伦多大学数据科学研究院) School of the Environment, University of Toronto(多伦多大学环境学院) David A. Dunlap Department of Astronomy and Astrophysics, University of Toronto(多伦多大学大卫·A·邓拉普天文与天体物理系) School of Public Health, Imperial College London(伦敦帝国学院公共卫生学院) Department of Astronomy, University of Washington(华盛顿大学天文学系)

AI总结 提出生成式代理框架,利用变分自编码器压缩Celerite先验,避免精确协方差运算,结合隐马尔可夫模型实现恒星耀斑的高效检测。

Comments Main paper: 19 pages, full paper: 34 pages. 4 appendices. 9 main figures, 21 figures in total. 4 tables. Poster Presenter, SSC 2026 (Statistical Society of Canada Annual Meeting) and ISBA 2026 (International Society for Bayesian Analysis World Meeting)

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.23517 2026-06-23 cs.LG stat.ML 新提交

Collapsed Effective Operators for Higher-order Structures

高阶结构的塌缩有效算子

Maximilian Krahn, Lennart Bastian, Vikas Garg, Björn Schuller, Tolga Birdal

机构 * Department of Computing, Imperial College London(帝国理工学院计算系) Aalto University(阿尔托大学) Chair of Health Informatics, Technical University of Munich(慕尼黑工业大学健康信息学教席) Munich Center for Machine Learning(慕尼黑机器学习中心) YaiYai Ltd(YaiYai有限公司)

AI总结 提出通过Schur补将高阶拉普拉斯算子塌缩为顶点级算子,保留正半定性并降低系统能量,在谱聚类、信号平滑和神经网络位置编码中提升性能。

Comments Accepted at ICML 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.23494 2026-06-23 cs.CV 新提交

Brain-Adapter: A Dual-Stream Vision-Language MIL Framework for Comprehensive 3D CT Diagnosis of Acute Intracranial Pathologies

Brain-Adapter: 一种用于急性颅内病理综合3D CT诊断的双流视觉-语言MIL框架

Zhenyu Yi, Zhiyun Song, Yusong Sun, Zelin Liu, Manman Fei, Zhenhao Li, Jiaxuan Zhao, Xu Han, Lichi Zhang

机构 * School of Biomedical Engineering, Shanghai Jiao Tong University(上海交通大学生物医学工程学院) Department of Computing, Imperial College London(伦敦帝国理工学院计算系)

AI总结 提出Brain-Adapter双流多实例学习框架,利用预训练2D生物医学视觉-语言模型和原始诊断报告,通过文本条件注意力和不确定性感知融合实现3D CT扫描的多标签分类,无需密集标注。

Comments Accepted to MICCAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.23473 2026-06-23 cs.CV 新提交

C^2GR: Coupled Comprehensive Generative Replay for a Continually Learnable Universal Segmentation Model

C^2GR: 用于持续学习通用分割模型的耦合综合生成回放

Wei Li, Jingyang Zhang, Guoan Wang, Junzhi Ning, Yang Chen, Guang Yang, Lixu Gu

机构 * Shanghai Jiao Tong University(上海交通大学) Southeast University(东南大学) Stevens Institute of Technology(史蒂文斯理工学院) Shanghai Artificial Intelligence Laboratory(上海人工智能实验室) Imperial College London(伦敦帝国学院)

AI总结 针对通用分割模型在任务增量学习中因图像外观和分割目标同时变化导致的遗忘问题,提出C^2GR框架,通过贝叶斯联合扩散和关系感知统一提示同步,合成历史任务图像-掩码对,缓解遗忘,性能仅下降2.44%。

Comments This paper has been submitted to a relevant journal

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.23391 2026-06-23 cs.LG cs.AI 新提交

Distribution-Aware Diffusion-LLM for Robust Ultra-Long-Term Time Series Forecasting

分布感知的扩散-大语言模型用于鲁棒的超长期时间序列预测

Falguni Ghosh, Vahid Hashemi, Bernhard Kainz

机构 * Friedrich-Alexander-Universität Erlangen-Nürnberg(埃尔朗根-纽伦堡大学) Imperial College London(伦敦帝国学院) AUDI AG(奥迪股份公司)

AI总结 提出Diffusion-LLM框架,将条件扩散模型集成到基于LLM的预测流程中,学习未来数据条件分布并改善语义对齐,在超长期和少样本预测中显著优于现有LLM基线。

Comments 18 pages, 6 figures, 8 tables. Accepted at 35th International Conference on Artificial Neural Networks (ICANN 2026)

详情

展开后加载摘要…

URL PDF HTML 收藏