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 1187
2512.03243 2025-12-04 stat.ML cs.LG math.PR math.ST stat.TH

Novelty detection on path space

路径空间上的新颖性检测

Ioannis Gasteratos, Antoine Jacquier, Maud Lemercier, Terry Lyons, Cristopher Salvi

机构 * Institute of Mathematics TU Berlin(柏林技术大学数学研究所) Department of Mathematics Imperial College London(伦敦帝国理工学院数学系) Mathematical Institute University of Oxford(牛津大学数学研究所)

AI总结 本文提出基于签名的检验统计量用于路径空间新颖性检测,通过运输成本不等式和洗牌积推导出CVaR替代公式,并设计新的单类SVM算法,评估了其在异常扩散和分子生物学数据中的性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2403.03631 2025-12-04 cs.LG cs.SY eess.SY

Marginalize, Rather than Impute: Probabilistic Wind Power Forecasting with Incomplete Data

边缘化而非填补:基于不完整数据的概率风功率预测

Honglin Wen, Pierre Pinson, Jie Gu, Zhijian Jin

机构 * School of Electrical Engineering, Shanghai Jiao Tong University(上海交通大学电气工程学院) Dyson School of Design Engineering, Imperial College London(帝国理工学院设计工程学院) Department of Technology, Management and Economics, Technical University of Denmark(丹麦技术大学技术、管理与经济系) Centre for Energy Research – CoRE, Aarhus University(奥胡斯大学能源研究中心)

AI总结 本文提出一种基于不完整数据的概率风功率预测方法,通过联合生成模型和边缘化技术提升预测质量并降低计算成本。

Comments Submitted to INFORMS Journal on Data Science

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.02740 2025-12-03 cs.LG cs.IT math.IT

Adversarial Jamming for Autoencoder Distribution Matching

对抗性干扰用于自编码器分布匹配

Waleed El-Geresy, Deniz Gündüz

机构 * Imperial College London(伦敦帝国学院)

AI总结 本文提出利用对抗性干扰作为辅助目标,实现自编码器潜在空间中高斯分布的匹配。

Comments Presented at ICASSP 2024. 5 pages, 3 figures

Journal ref ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Seoul, Republic of Korea, 2024, pp. 7605-7609

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.26012 2025-12-03 cs.AI

AutoSurvey2: Empowering Researchers with Next Level Automated Literature Surveys

AutoSurvey2:赋能研究人员的下一代自动化文献综述

Siyi Wu, Chiaxin Liang, Ziqian Bi, Leyi Zhao, Tianyang Wang, Junhao Song, Yichao Zhang, Keyu Chen, Benji Peng, Xinyuan Song

机构 * University of Texas at Arlington(德克萨斯理工大学) AI Agent Lab(人工智能代理实验室) Indiana University(印第安纳大学) Ohio State University(俄亥俄州立大学) Imperial College London(伦敦帝国理工学院) Georgia Institute of Technology(佐治亚理工学院) Appcubic(Appcubic公司) Emory University(埃默里大学)

AI总结 AutoSurvey2通过多阶段流程实现自动化文献综述生成,结合检索增强合成与结构化评估,提升综述的连贯性与相关性,为学术写作提供可扩展解决方案。

Comments TKDD 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.02502 2025-12-03 cs.IR cs.AI

AskNearby: An LLM-Based Application for Neighborhood Information Retrieval and Personalized Cognitive-Map Recommendations

AskNearby: 一种基于LLM的邻里信息检索与个性化认知地图推荐应用

Luyao Niu, Zhicheng Deng, Boyang Li, Nuoxian Huang, Ruiqi Liu, Wenjia Zhang

机构 * Peking University(北京大学) Qianmo Smart Link(千摩智能链接) New York University(纽约大学) Imperial College London(伦敦帝国学院) Tencent(腾讯) Tongji University(同济大学)

AI总结 AskNearby通过整合检索增强生成和认知地图模型,提升邻里信息检索与个性化推荐效果,解决局部生活信息可及性问题。

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.01463 2025-12-03 cs.AR cs.LG hep-ex

hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware

hls4ml:一种灵活、开源的深度学习在可重构硬件上加速平台

Jan-Frederik Schulte, Benjamin Ramhorst, Chang Sun, Jovan Mitrevski, Nicolò Ghielmetti, Enrico Lupi, Dimitrios Danopoulos, Vladimir Loncar, Javier Duarte, David Burnette, Lauri Laatu, Stylianos Tzelepis, Konstantinos Axiotis, Quentin Berthet, Haoyan Wang, Paul White, Suleyman Demirsoy, Marco Colombo, Thea Aarrestad, Sioni Summers, Maurizio Pierini, Giuseppe Di Guglielmo, Jennifer Ngadiuba, Javier Campos, Ben Hawks, Abhijith Gandrakota, Farah Fahim, Nhan Tran, George Constantinides, Zhiqiang Que, Wayne Luk, Alexander Tapper, Duc Hoang, Noah Paladino, Philip Harris, Bo-Cheng Lai, Manuel Valentin, Ryan Forelli, Seda Ogrenci, Lino Gerlach, Rian Flynn, Mia Liu, Daniel Diaz, Elham Khoda, Melissa Quinnan, Russell Solares, Santosh Parajuli, Mark Neubauer, Christian Herwig, Ho Fung Tsoi, Dylan Rankin, Shih-Chieh Hsu, Scott Hauck

机构 * Purdue University(普渡大学) ETH Zurich(苏黎世联邦理工学院) California Institute of Technology(加州理工学院) Fermi National Accelerator Lab(费米国家加速器实验室) European Organization for Nuclear Research (CERN)(欧洲核子研究中心) University of California San Diego(加州大学圣地亚哥分校) Imperial College London(伦敦帝国理工学院) National Technical University of Athens(希腊国家技术大学) University of Geneva(日内瓦大学) Altera Corporation(阿尔特拉公司) Discovery Partners Institute(发现伙伴研究所) Massachusetts Institute of Technology(麻省理工学院) National Yang Ming Chiao Tung University(国立阳明交通大学) Northwestern University(西北大学) Princeton University(普林斯顿大学) University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校) University of Michigan(密歇根大学) University of Pennsylvania(宾夕法尼亚大学) University of Washington(华盛顿大学)

AI总结 hls4ml是一种开源平台,用于将深度学习模型转换为可重构硬件上的HLS代码,以实现低延迟、低资源消耗的ML推理加速。

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.01913 2025-12-02 eess.IV cs.CV

Disentangling Progress in Medical Image Registration: Beyond Trend-Driven Architectures towards Domain-Specific Strategies

解构医学图像配准的进步:超越趋势驱动的架构,迈向领域特定的策略

Bailiang Jian, Jiazhen Pan, Rohit Jena, Morteza Ghahremani, Hongwei Bran Li, Daniel Rueckert, Christian Wachinger, Benedikt Wiestler

机构 * Technical University of Munich(慕尼黑技术大学) Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心) Imperial College London(伦敦帝国理工学院) University of Pennsylvania(宾夕法尼亚大学) National University of Singapore(新加坡国立大学)

AI总结 本文通过模块化框架解构医学图像配准中趋势驱动架构与领域特定设计的影响,发现后者在性能提升上优于前者,推动研究重点转向领域特定原则。

Comments Submitted to Medical Image Analysis. Journal Extension of arXiv:2407.19274

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.01908 2025-12-02 cs.CV

SARL: Spatially-Aware Self-Supervised Representation Learning for Visuo-Tactile Perception

SARL: 用于视觉-触觉感知的时空自监督表征学习

Gurmeher Khurana, Lan Wei, Dandan Zhang

机构 * Department of Bioengineering, Imperial-X Initiative, Imperial College London(生物工程系、Imperial-X计划、帝国理工学院伦敦分校)

AI总结 SARL通过引入三个地图级目标,提升视觉-触觉数据的自监督学习效果,实现更精确的机器人感知。

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.01626 2025-12-02 cs.SD cs.NE

Parallel Delayed Memory Units for Enhanced Temporal Modeling in Biomedical and Bioacoustic Signal Analysis

并行延迟记忆单元用于生物医学和生物声学信号分析中的增强时间建模

Pengfei Sun, Wenyu Jiang, Paul Devos, Dick Botteldooren

机构 * Department of Information Technology, WAVES Research Group, Ghent University(信息科技系、WAVES研究组、根特大学) Department of Electrical and Electronic Engineering, Imperial College London(电子与电气工程系、伦敦帝国学院)

AI总结 并行延迟记忆单元通过门控延迟线机制提升生物医学和生物声学信号的时间建模能力,增强记忆效率与模型性能。

Comments Accepted for publication in IEEE Transactions on Audio, Speech and Language Processing, 2025

Journal ref IEEE Transactions on Audio, Speech and Language Processing, 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.00428 2025-12-02 cs.CV

Recognizing Pneumonia in Real-World Chest X-rays with a Classifier Trained with Images Synthetically Generated by Nano Banana

用Nano Banana生成的合成图像训练分类器以在真实世界胸片中识别肺炎

Jiachuan Peng, Kyle Lam, Jianing Qiu

机构 * Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学) University of Oxford(牛津大学) Imperial College London(伦敦帝国学院)

AI总结 利用Nano Banana生成的合成图像训练分类器,实现在真实世界胸片中识别肺炎,取得较高AUROC和AUPR,但存在提示设计和数据对齐的挑战。

Comments 9 pages

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.16671 2025-12-02 cs.CL cs.AI cs.CV

MimeQA: Towards Socially-Intelligent Nonverbal Foundation Models

MimeQA: 向具有社会智能的非语言基础模型迈进

Hengzhi Li, Megan Tjandrasuwita, Yi R. Fung, Armando Solar-Lezama, Paul Pu Liang

机构 * Massachusetts Institute of Technology(麻省理工学院) Imperial College London(伦敦帝国理工学院)

AI总结 MimeQA通过引入非语言互动数据集,评估视频大语言模型在非语言社会推理中的表现,发现其准确率较低,人类表现更优。

Comments NeurIPS 2025 Datasets and Benchmarks

详情

展开后加载摘要…

URL PDF HTML 收藏
2501.16086 2025-12-02 stat.ML cs.LG

Value-oriented forecast reconciliation for renewables in electricity markets

面向价值的可再生能源电力市场预测协调

Honglin Wen, Pierre Pinson

机构 * Department of Electrical Engineering, Shanghai Jiao Tong University, China(上海交通大学电气工程系) Dyson School of Design Engineering, Imperial College London, United Kingdom(帝国理工学院伦敦戴森设计工程学院) Department of Technology, Management and Economics, Technical University of Denmark, Denmark(丹麦技术大学技术、管理与经济系) CoRE, Aarhus University, Denmark(奥胡斯大学CoRE)

AI总结 本文提出了一种面向价值的预测协调方法,通过纳什谈判框架确保公平性,提升多智能体环境下可再生能源交易的利润分配效率。

Comments preprint of European Journal of Operational Research

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.00077 2025-12-02 cs.RO

A Hierarchical Framework for Humanoid Locomotion with Supernumerary Limbs

具有冗余肢体的人形机器人类似运动的分层框架

Bowen Zhi

机构 * Imperial College London(伦敦帝国学院)

AI总结 本文提出了一种分层框架,通过结合学习运动和模型平衡,提高人形机器人在冗余肢体下的运动稳定性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.00015 2025-12-02 cs.HC cs.AI

The Impact of Concept Explanations and Interventions on Human-Machine Collaboration

概念解释与干预对人机协作的影响

Jack Furby, Dan Cunnington, Dave Braines, Alun Preece

机构 * Cardiff University(卡迪夫大学) Imperial College London(伦敦帝国学院) IBM Research Europe(IBM欧洲研究院)

AI总结 本文研究了概念瓶颈模型(CBMs)在人机协作中的影响,发现其提升可解释性但未显著提高任务准确率。

Comments 24 pages, 5 figures, 8 tables. Accepted at The World Conference on eXplainable Artificial Intelligence 2025 (XAI-2025). The Version of Record of this chapter is published in Explainable Artificial Intelligence, and is available online at https://doi.org/10.1007/978-3-032-08317-3_12. The version published here includes minor typographical corrections

Journal ref Explainable Artificial Intelligence, Springer Nature Switzerland, 2026, pp. 255-280

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.22483 2025-12-01 cs.LG

Enhancing Trustworthiness with Mixed Precision: Benchmarks, Opportunities, and Challenges

通过混合精度提升可信度:基准测试、机遇与挑战

Guanxi Lu, Hao Mark Chen, Zhiqiang Que, Wayne Luk, Hongxiang Fan

机构 * Department of Computing Imperial College London(计算系 帝国理工学院伦敦分校)

AI总结 本文研究了量化对可信度指标的影响,提出了一种混合精度集合投票方法,提升了可信度指标性能。

Comments ASP-DAC 2026 Special Session

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.18615 2025-12-01 cs.LG stat.ML

Bayesian-based Online Label Shift Estimation with Dynamic Dirichlet Priors

基于贝叶斯的在线标签偏移估计与动态Dirichlet先验

Jiawei Hu, Javier A. Barria

机构 * Department of Electrical and Electronic Engineering, Imperial College London(帝国理工学院伦敦分校电子与电气工程系)

AI总结 本文提出FMAPLS和online-FMAPLS方法,通过动态优化Dirichlet超参数和类别先验,有效解决标签偏移问题,提升分类性能。

Comments 13 pages, submitted to IEEE journal for possible publication

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.07619 2025-12-01 cs.LG q-bio.QM

The Catechol Benchmark: Time-series Solvent Selection Data for Few-shot Machine Learning

儿茶酚基准:用于少样本机器学习的时序溶剂选择数据

Toby Boyne, Juan S. Campos, Becky D. Langdon, Jixiang Qing, Yilin Xie, Shiqiang Zhang, Calvin Tsay, Ruth Misener, Daniel W. Davies, Kim E. Jelfs, Sarah Boyall, Thomas M. Dixon, Linden Schrecker, Jose Pablo Folch

机构 * Department of Computing, Imperial College London(计算系,帝国理工学院伦敦分校) Department of Chemistry, Imperial College London(化学系,帝国理工学院伦敦分校) SOLVE Chemistry(SOLVE化学)

AI总结 本文提出了一种用于少样本机器学习的时序溶剂选择数据集,通过大规模连续工艺条件样本,挑战机器学习模型,应用于溶剂替代和可持续制造。

Comments 10 pages main, 22 pages total, 8 figures, 7 tables. Accepted to NeurIPS Datasets and Benchmarks track 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.21283 2025-11-27 math.DS cs.LG

On the Periodic Orbits of the Dual Logarithmic Derivative Operator

关于双对数导数算子的周期轨道

Xiaohang Yu, William Knottenbelt

机构 * Imperial College London(伦敦帝国学院)

AI总结 本文研究了双对数导数算子的周期轨道,通过分类非退化周期-2解和不动点,揭示了算子在函数空间上的动力学结构。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.20865 2025-11-27 cs.CV

Estimating Fog Parameters from a Sequence of Stereo Images

从立体图像序列中估计雾参数

Yining Ding, João F. C. Mota, Andrew M. Wallace, Sen Wang

机构 * Edinburgh Centre for Robotics, the School of Mathematical and Computer Sciences, Heriot-Watt University(爱丁堡机器人中心、数学与计算机科学学院、赫瑞斯泰大学) School of Engineering and Physical Sciences, Heriot-Watt University(工程与物理科学学院、赫瑞斯泰大学) Sense Robotics Lab, Department of Electrical and Electronic Engineering, Imperial College London(感知机器人实验室、电子与电气工程系、伦敦帝国学院)

AI总结 本文提出了一种从立体图像序列中同时估计雾参数的方法,并创建了新的SDIRF数据集,用于评估算法在雾环境下的性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2504.16136 2025-11-27 cs.LG

Active Learning Methods for Efficient Data Utilization and Model Performance Enhancement

主动学习方法用于高效数据利用和模型性能提升

Chiung-Yi Tseng, Junhao Song, Ziqian Bi, Tianyang Wang, Chia Xin Liang, Xinyuan Song, Ming Liu

机构 * AI Agent Lab(AI代理实验室) Imperial College London(伦敦帝国学院) Purdue University(普渡大学) University of Liverpool(利物浦大学) JTB Technology Corp.(JTB科技公司) Emory University(埃默里大学)

AI总结 本文提出主动学习方法,通过减少标注数据需求提升模型性能,探讨其在多个领域的应用及挑战。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.20116 2025-11-26 cs.CV cs.AI

LungEvaty: A Scalable, Open-Source Transformer-based Deep Learning Model for Lung Cancer Risk Prediction in LDCT Screening

LungEvaty: 一种可扩展、开源的基于Transformer的深度学习模型,用于LDCT筛查中的肺癌风险预测

Johannes Brandt, Maulik Chevli, Rickmer Braren, Georgios Kaissis, Philip Müller, Daniel Rueckert

机构 * Technical University of Munich (TUM)(慕尼黑技术大学) TUM University Hospital(慕尼黑技术大学医院) Imperial College London(伦敦帝国学院) Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心)

AI总结 LungEvaty是一种基于Transformer的深度学习模型,利用LDCT扫描预测肺癌风险,通过全肺输入和解剖学指导注意力实现高效且准确的预测。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.19431 2025-11-26 cs.CV physics.ao-ph

Cloud4D: Estimating Cloud Properties at a High Spatial and Temporal Resolution

Cloud4D: 高空间和时间分辨率下云特性估计

Jacob Lin, Edward Gryspeerdt, Ronald Clark

机构 * Department of Computer Science University of Oxford(计算机科学系牛津大学) Department of Physics Imperial College London(物理系伦敦帝国学院)

AI总结 Cloud4D通过同步地面相机和2D-3D变换器,实现了高空间和时间分辨率的云特性估计,提升空间-时间分辨率并保持高精度。

Comments NeurIPS 2025 Spotlight, project page: https://cloud4d.jacob-lin.com/

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.24160 2025-11-26 cs.LG

Identifiable learning of dissipative dynamics

可识别性学习耗散动力学

Aiqing Zhu, Beatrice W. Soh, Grigorios A. Pavliotis, Qianxiao Li

机构 * Department of Mathematics, National University of Singapore(新加坡国立大学数学系) Department of Chemical and Biomolecular Engineering, National University of Singapore(新加坡国立大学化学与生物分子工程系) Department of Mathematics, Imperial College London(伦敦帝国理工学院数学系) Institute for Functional Intelligent Materials, National University of Singapore(新加坡国立大学功能智能材料研究所)

AI总结 本文提出了一种可识别的神经框架,用于直接从轨迹学习耗散随机动力学,并提供非平衡动力学的原理性度量。

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.01196 2025-11-26 cs.LG cs.AI cs.ET cs.HC

Are Large Brainwave Foundation Models Capable Yet? Insights from Fine-tuning

大规模脑电基础模型是否已具备能力?来自微调的洞察

Na Lee, Konstantinos Barmpas, Yannis Panagakis, Dimitrios Adamos, Nikolaos Laskaris, Stefanos Zafeiriou

机构 * Imperial College London(伦敦帝国学院) Archimedes / Athena Research Unit(阿基米德/雅典娜研究单位) Aristotle University of Thessaloniki(雅典娜大学) Kapodistrian University of Athens(雅典kapodistrian大学)

AI总结 本文通过微调实验评估了大规模脑电基础模型的能力,发现其在BCI任务中效率有限,提出LoRA技术可提升性能,强调需重新设计架构以提升脑电分析效果。

Journal ref International Conference on Machine Learning (ICML) 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.16383 2025-11-26 cs.CL

Large Language Models in Argument Mining: A Survey

大型语言模型在论证挖掘中的应用:综述

Hao Li, Viktor Schlegel, Yizheng Sun, Riza Batista-Navarro, Goran Nenadic

机构 * University of Manchester(曼彻斯特大学) Imperial College London(伦敦帝国学院)

AI总结 本文综述了大型语言模型对论证挖掘领域的影响,分析了LLM如何改变任务设计、数据集构建和评估方法,并提出了未来研究方向。

Comments Work draft

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.19165 2025-11-25 eess.IV cs.CV

Extreme Cardiac MRI Analysis under Respiratory Motion: Results of the CMRxMotion Challenge

极端呼吸运动下的心脏MRI分析:CMRxMotion挑战结果

Kang Wang, Chen Qin, Zhang Shi, Haoran Wang, Xiwen Zhang, Chen Chen, Cheng Ouyang, Chengliang Dai, Yuanhan Mo, Chenchen Dai, Xutong Kuang, Ruizhe Li, Xin Chen, Xiuzheng Yue, Song Tian, Alejandro Mora-Rubio, Kumaradevan Punithakumar, Shizhan Gong, Qi Dou, Sina Amirrajab, Yasmina Al Khalil, Cian M. Scannell, Lexiaozi Fan, Huili Yang, Xiaowu Sun, Rob van der Geest, Tewodros Weldebirhan Arega, Fabrice Meriaudeau, Caner Özer, Amin Ranem, John Kalkhof, İlkay Öksüz, Anirban Mukhopadhyay, Abdul Qayyum, Moona Mazher, Steven A Niederer, Carles Garcia-Cabrera, Eric Arazo, Michal K. Grzeszczyk, Szymon Płotka, Wanqin Ma, Xiaomeng Li, Rongjun Ge, Yongqing Kou, Xinrong Chen, He Wang, Chengyan Wang, Wenjia Bai, Shuo Wang

机构 * Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, Shanghai, Shanghai 200032, China Shanghai Key Laboratory of MICCAI, Fudan University, Shanghai, Shanghai 200032, China Department of Electrical Electronic Engineering \& I-X, Imperial College London, London, London SW7 2AZ, United Kingdom Department of Radiology, Zhongshan Hospital Affiliated to Fudan University, Shanghai, Shanghai 200032, China Department of Computing, Imperial College London, London, London SW7 2AZ, United Kingdom School of Computer Science, University of Sheffield, Sheffield, S1 4DP, United Kingdom Department of Engineering Science, University of Oxford, Oxford, OX2 0ES, United Kingdom Shanghai Pudong Hospital Human Phenome Institute, Fudan University, Shanghai, 201203, China School of Computer Science, University of Nottingham, Nottingham, NG8 1BB, United Kingdom Diagnostic Imaging, University of Alberta, Edmonton, AB T6G 1K4, Canada Department of Computer Science Engineering, The Chinese University of Hong Kong, Hong Kong, Hong Kong 000000, China The D-Lab, Department of Precision Medicine, GROW - Research Institute for Oncology Reproduction, Maastricht University, 6220 MD Maastricht, The Netherlands Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven 5612 AZ, The Netherlands Department of Radiology, Northwestern University, 737 N. Michigan Ave, Suite 1600, Chicago 60611, United States United Imaging Research, 393 Middle Huaxia Road, Pudong, Shanghai 201210, China Division of Image Processing, Department of Radiology, Leiden University Medical Center, PO Box 9600, Leiden 2300 RC, The Netherlands Université Bourgogne Europe, CNRS, ICMUB UMR 6302, 21000 Dijon, France Istanbul Technical University, Maslak, 34467, İstanbul, Türkiye Computer Science, Technical University of Darmstadt, Karolinenpl. 5, 64289 Darmstadt, Germany Lung Institute, Faculty of Medicine, Imperial College London, Guy Scadding Building, Cale Street, London, SW3 6LY,United Kingdom Hawkes Institute, Department of Computer Science, University College London, 66-72 Gower St, London, United Kingdom School of Medicine, University College Dublin, Belfield, Dublin, D04 V1W8, Ireland CeADAR: Ireland's Centre for AI, University College Dublin, Belfield, Dublin, D04 V1W8, Ireland Sano Centre for Computational Medicine, Czarnowiejska 36, 30-054, Krakow, Poland Faculty of Mathematics Computer Science, Jagiellonian University, S. Łojasiewicza 6, Krakow, Poland Department of Electronic Computer Engineering, The Hong Kong University of Science School of Instrument Science Engineering, Southeast University, Nanjing, Nanjing 210096, China College of Artificial Intelligence, Nanjing University of Aeronautics Academy for Engineering Technology, Fudan University, Shanghai, Shanghai 200433, China College of Biomedical Engineering, Fudan University, Shanghai, Shanghai 200433, China Institute of Science Technology for Brain-inspired Intelligence, Fudan University, Shanghai, Shanghai 200433, China Department of Brain Sciences, Imperial College London, London, London SW7 2AZ, United Kingdom Data Science Institute, Imperial College London, London, London SW7 2AZ, United Kingdom

AI总结 本文提出CMRxMotion挑战,通过公开数据集评估深度学习模型在呼吸运动干扰下的心脏MRI分析性能,并探讨运动伪影对临床生物标志物的影响。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.16567 2025-11-24 cs.CV

POMA-3D: The Point Map Way to 3D Scene Understanding

POMA-3D:点地图方式的3D场景理解

Ye Mao, Weixun Luo, Ranran Huang, Junpeng Jing, Krystian Mikolajczyk

机构 * Imperial College London(伦敦帝国学院)

AI总结 POMA-3D通过点地图实现3D场景理解,结合自监督学习和多视角对齐策略,提升3D任务表现。

Comments 11 pages, 6 tables, 5 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.16602 2025-11-21 cs.AI

Bridging VLMs and Embodied Intelligence with Deliberate Practice Policy Optimization

通过刻意练习策略优化弥合视觉语言模型与具身智能之间的鸿沟

Yi Zhang, Che Liu, Xiancong Ren, Hanchu Ni, Yingji Zhang, Shuai Zhang, Zeyuan Ding, Jiayu Hu, Haozhe Shan, Junbo Qi, Yan Bai, Dengjie Li, Jiachen Luo, Yidong Wang, Yong Dai, Zenglin Xu, Bin Shen, Qifan Wang, Jian Tang, Xiaozhu Ju

机构 * X-Humanoid Imperial College London(帝国理工学院) Peking University(北京大学) University of Manchester(曼彻斯特大学) Westlake University(西湖大学) Fudan University(复旦大学) Waseda University(早稻田大学) Nvidia Queen Mary University of London(伦敦大学玛丽女王学院) Celonis AI Meta AI

AI总结 本文提出DPPO框架,通过监督微调与强化学习交替训练,提升具身智能系统在稀疏数据下的学习效率,并在性能和参数规模上取得显著优势。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.16494 2025-11-21 cs.CV cs.AI

Physics-Informed Machine Learning for Efficient Sim-to-Real Data Augmentation in Micro-Object Pose Estimation

基于物理的机器学习用于微物体位姿估计中的高效仿真到现实数据增强

Zongcai Tan, Lan Wei, Dandan Zhang

机构 * Department of Bioengineering, Imperial-X Initiative, Imperial College London(生物工程系、Imperial-X计划、帝国理工学院伦敦分校)

AI总结 本文提出基于物理的深度生成学习框架,通过整合波动光学和深度对齐,高效生成高保真显微镜图像,提升微机器人位姿估计的精度与效率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.14894 2025-11-21 cs.LG hep-ex

Leveraging Reinforcement Learning, Genetic Algorithms and Transformers for background determination in particle physics

利用强化学习、遗传算法和变换器进行粒子物理中背景确定

Guillermo Hijano Mendizabal, Davide Lancierini, Alex Marshall, Andrea Mauri, Patrick Haworth Owen, Mitesh Patel, Konstantinos Petridis, Shah Rukh Qasim, Nicola Serra, William Sutcliffe, Hanae Tilquin

机构 * Physik-Institut, Universität Zürich(苏黎世大学物理研究所) Imperial College London(帝国理工学院伦敦校区) H.H. Wills Physics Laboratory, University of Bristol(布里斯托大学H.H.威尔斯物理实验室) University of Bristol(布里斯托大学)

AI总结 本文提出利用强化学习、遗传算法和变换器解决粒子物理中背景确定问题,通过系统方法提升衰变测量精度。

Comments 34 pages, 12 figures

详情

展开后加载摘要…

URL PDF HTML 收藏