重大更新！公众号每日速递覆盖arXiv所有方向,涵盖CS|物理|数学|经济|统计|金融|生物|电气等领域。点击查看历史文章列表，可在公众号内订阅#arXiv每日学术速递话题，接收每日推送。点击访问arxivdaily.com每日更新更及时。

人工智能cs.AI 方向，今日共计43篇

【1】 Do Vision-Language Pretrained Models Learn Primitive Concepts?
标题：视觉语言预先训练的模型学习原始概念吗？
作者：Tian Yun,Usha Bhalla,Ellie Pavlick,Chen Sun
机构*：Brown University
备注：Under review
链接：点击下载PDF文件

【2】 A Closer Look at Rehearsal-Free Continual Learning
标题：细看排练--免费的持续学习
作者：James Seale Smith,Junjiao Tian,Yen-Chang Hsu,Zsolt Kira
机构*：Georgia Institute of Technology,Samsung Research America
链接：点击下载PDF文件

【3】 R2L: Distilling Neural Radiance Field to Neural Light Field for Efficient Novel View Synthesis
标题：R2L：提取神经辐射场到神经光场以实现高效的新视点合成
作者：Huan Wang,Jian Ren,Zeng Huang,Kyle Olszewski,Menglei Chai,Yun Fu,Sergey Tulyakov
机构*：Northeastern University, Snap Inc.
备注：Project: this https URL
链接：点击下载PDF文件

【4】 LEAD1.0: A Large-scale Annotated Dataset for Energy Anomaly Detection in Commercial Buildings
标题：LEAD1.0：一种用于商业建筑能耗异常检测的大规模注释数据集
作者：Manoj Gulati,Pandarasamy Arjunan
机构*：Singapore Management University, Berkeley Education Alliance for, Research in Singapore Limited
链接：点击下载PDF文件

【5】 Continuous Scene Representations for Embodied AI
标题：面向具体化人工智能的连续场景表示
作者：Samir Yitzhak Gadre,Kiana Ehsani,Shuran Song,Roozbeh Mottaghi
机构*： Columbia University, Allen Institute for AI, University of Washington, prior.allenai.orgprojectscsr
备注：CVPR 2022
链接：点击下载PDF文件

【6】 Generation and Simulation of Synthetic Datasets with Copulas
标题：基于Copulas的合成数据集的生成与仿真
作者：Regis Houssou,Mihai-Cezar Augustin,Efstratios Rappos,Vivien Bonvin,Stephan Robert-Nicoud
机构*： universities and the private sector
链接：点击下载PDF文件

【7】 Dual Temperature Helps Contrastive Learning Without Many Negative Samples: Towards Understanding and Simplifying MoCo
标题：双温有助于无负样本对比学习：对MoCo的理解和简化
作者：Chaoning Zhang,Kang Zhang,Trung X. Pham,Axi Niu,Zhinan Qiao,Chang D. Yoo,In So Kweon
机构*： KAIST, Northwestern Polytechnical University, University of North Texas
备注：Accepted by CVPR2022
链接：点击下载PDF文件

【8】 VL-InterpreT: An Interactive Visualization Tool for Interpreting Vision-Language Transformers
标题：VL-Interpret：用于解释视觉语言转换器的交互式可视化工具
作者：Estelle Aflalo,Meng Du,Shao-Yen Tseng,Yongfei Liu,Chenfei Wu,Nan Duan,Vasudev Lal
机构*：Intel Labs, UCLA, Microsoft Research
备注：CVPR 2022 demo track
链接：点击下载PDF文件

【9】 3D Equivariant Graph Implicit Functions
标题：三维等变图隐函数
作者：Yunlu Chen,Basura Fernando,Hakan Bilen,Matthias Niener,Efstratios Gavves
机构*： University of Amsterdam, CFAR, IHPC, ASTAR, University of Edinburgh, Technical University of Munich
备注：Video: this https URL
链接：点击下载PDF文件

【10】 Adaptiv Mean-Residue Loss for Robust Facial Age Estimation
标题：基于自适应均值残差损失的人脸年龄稳健估计
作者：Ziyuan Zhao,Peisheng Qian,Yubo Hou,Zeng Zeng
机构*：Institute of Infocomm Research (I,R), ASTAR, Singapore, Artificial Intelligence, Analytics And Informatics (AI,), ASTAR, Singapore, School of Computer Science and Engineering, Nanyang Technological University, Singapore
备注：Accepted by IEEE International Conference on Multimedia and Expo (ICME 2022)
链接：点击下载PDF文件

【11】 Imitate and Repurpose: Learning Reusable Robot Movement Skills From Human and Animal Behaviors
标题：模仿和再利用：从人和动物的行为中学习可重复使用的机器人运动技能
作者：Steven Bohez,Saran Tunyasuvunakool,Philemon Brakel,Fereshteh Sadeghi,Leonard Hasenclever,Yuval Tassa,Emilio Parisotto,Jan Humplik,Tuomas Haarnoja,Roland Hafner,Markus Wulfmeier,Michael Neunert,Ben Moran,Noah Siegel,Andrea Huber,Francesco Romano,Nathan Batchelor,Federico Casarini,Josh Merel,Raia Hadsell,Nicolas Heess
机构*：DeepMind, London, UK
备注：30 pages, 9 figures, 8 tables, 14 videos at this https URL , submitted to Science Robotics
链接：点击下载PDF文件

【12】 A Rich Recipe Representation as Plan to Support Expressive Multi Modal Queries on Recipe Content and Preparation Process
标题：一种丰富的食谱表示为平面，支持对食谱内容和制作过程的可表达的多模式查询
作者：Vishal Pallagani,Priyadharsini Ramamurthy,Vedant Khandelwal,Revathy Venkataramanan,Kausik Lakkaraju,Sathyanarayanan N. Aakur,Biplav Srivastava
机构*：AI Institute, University of South Carolina and Oklahoma State University
链接：点击下载PDF文件

【13】 Interpretation of Black Box NLP Models: A Survey
标题：黑箱NLP模型的解释：综述
作者：Shivani Choudhary,Niladri Chatterjee,Subir Kumar Saha
机构*：Indian Institute of Technology Delhi, Hauz Khas, Delhi-, India
链接：点击下载PDF文件

【14】 Scientific and Technological Text Knowledge Extraction Method of based on Word Mixing and GRU
标题：一种基于词语混合和GRU的科技文本知识抽取方法
作者：Suyu Ouyang,Yingxia Shao,Junping Du,Ang Li
机构*：Scientific and Technological Text Knowledge Extraction Method of based, on Word Mixing and GRU, （College of Computer Scienc, Beijing Key Laboratory of Intelligent Telecommunication Software and Multimedia, Beijing ,）
备注：8 pages,2 figures
链接：点击下载PDF文件

【15】 IITD-DBAI: Multi-Stage Retrieval with Pseudo-Relevance Feedback and Query Reformulation
标题：IITD-DBAI：带伪相关反馈和查询重写的多阶段检索
作者：Shivani Choudhary
机构*：Indian Institute of Technology, Delhi
链接：点击下载PDF文件

【16】 Human Instance Segmentation and Tracking via Data Association and Single-stage Detector
标题：基于数据关联和单级检测器的人体实例分割与跟踪
作者：Lu Cheng,Mingbo Zhao
机构*： Zhao are with the School of Information Sci-ence and Technology, Donghua University
链接：点击下载PDF文件

【17】 MBORE: Multi-objective Bayesian Optimisation by Density-Ratio Estimation
标题：基于密度比估计的多目标贝叶斯优化
作者：George De Ath,Tinkle Chugh,Alma A. M. Rahat
机构*：Department of Computer Science, University of Exeter, Exeter, United Kingdom, Swansea University, Swansea, United Kingdom
备注：To appear at Genetic and Evolutionary Computation Conference (GECCO '22). 10 pages (main paper) + 25 pages (supplement). Code avaliable at this https URL
链接：点击下载PDF文件

【18】 A survey of neural models for the automatic analysis of conversation: Towards a better integration of the social sciences
标题：会话自动分析的神经模型综述：走向社会科学的更好融合
作者：Chloé Clavel,Matthieu Labeau,Justine Cassell
机构*：LTCI, Telecom-Paris, Institut Polytechnique de Paris, France, School of Computer Science, Carnegie Mellon University, USA, Inria Paris, France
链接：点击下载PDF文件

【19】 MPS-NeRF: Generalizable 3D Human Rendering from Multiview Images
标题：MPS-NERF：多视点图像的通用化三维人体绘制
作者：Xiangjun Gao,Jiaolong Yang,Jongyoo Kim,Sida Peng,Zicheng Liu,Xin Tong
机构*：Beijing Institute of Technology, Microsoft Research Asia, Zhejiang University
链接：点击下载PDF文件

【20】 Ransomware Detection using Process Memory
标题：使用进程内存的勒索软件检测
作者：Avinash Singh,Richard Adeyemi Ikuesan,Hein Venter
机构*：University of Pretoria, South Africa, Community College Qatar, Doha, Qatar
Journal-ref：17th International Conference on Cyber Warfare and Security, 032022
链接：点击下载PDF文件

【21】 Rethinking Portrait Matting with Privacy Preserving
标题：对肖像画中隐私保护的再思考
作者：Sihan Ma,Jizhizi Li,Jing Zhang,He Zhang,Dacheng Tao
机构*：Received: date Accepted: date
备注：An extended version of the ACM Multimedia 2021 paper "Privacy-Preserving Portrait Matting". The code, dataset, and models are available at this https URL arXiv admin note: substantial text overlap with arXiv:2104.14222
链接：点击下载PDF文件

【22】 TrajGen: Generating Realistic and Diverse Trajectories with Reactive and Feasible Agent Behaviors for Autonomous Driving
标题：TrajGen：为自动驾驶生成具有反应性和可行性的智能体行为，生成逼真和多样的轨迹
作者：Qichao Zhang,Yinfeng Gao,Yikang Zhang,Youtian Guo,Dawei Ding,Yunpeng Wang,Peng Sun,Dongbin Zhao
机构*： University of Chinese Academy of Sciences
链接：点击下载PDF文件

【23】 Mask Atari for Deep Reinforcement Learning as POMDP Benchmarks
标题：掩码Atari用于深度强化学习作为POMDP基准
作者：Yang Shao,Quan Kong,Tadayuki Matsumura,Taiki Fuji,Kiyoto Ito,Hiroyuki Mizuno
机构*：Mizuno, Hitachi Ltd., Higashi koigakubo ,-, Kokubunji-shi, Tokyo, Japan
链接：点击下载PDF文件

【24】 Bangla hate speech detection on social media using attention-based recurrent neural network
标题：基于注意力的递归神经网络在社交媒体上检测孟加拉仇恨语音
作者：Amit Kumar Das,Abdullah Al Asif,Anik Paul,Md. Nur Hossain
机构*：Bangla hate speech detection on social, media using attention-based recurrent, neural network
Journal-ref：Type: Journal Language: English Publisher: De Gruyter First published: Setember 1, 1991 Publication Frequency: 1 Issue per Year Audience: researchers in the field of intelligent systems
链接：点击下载PDF文件

【25】 Learning Decoupling Features Through Orthogonality Regularization
标题：用正交化正则化学习解耦特征
作者：Li Wang,Rongzhi Gu,Weiji Zhuang,Peng Gao,Yujun Wang,Yuexian Zou
机构*： ADSPLAB, School of ECE, Peking University, Shenzhen, China, Xiaomi Inc., Beijing, China
备注：Accepted at ICASSP 2022
链接：点击下载PDF文件

【26】 ReSTR: Convolution-free Referring Image Segmentation Using Transformers
标题：RESTR：基于变换的无卷积参考图像分割
作者：Namyup Kim,Dongwon Kim,Cuiling Lan,Wenjun Zeng,Suha Kwak
机构*：POSTECH, Microsoft Research Asia, EIT Institute for Advanced Study
备注：CVPR 2022 accepted
链接：点击下载PDF文件

【27】 An unsupervised cluster-level based method for learning node representations of heterogeneous graphs in scientific papers
标题：一种基于无监督聚类的科技论文异质图节点表示学习方法
作者：Jie Song,Meiyu Liang,Zhe Xue,Junping Du,Kou Feifei
机构*：An unsupervised cluster-level based method for learning node, representations of heterogeneous graphs in scientific papers, (Beijing Key Laboratory of Intelligent Communication Software and Multimedia, School of Computer, Beijing University of Posts and
备注：10 pages,3 pages
链接：点击下载PDF文件

【28】 Generation of Speaker Representations Using Heterogeneous Training Batch Assembly
标题：基于异质训练批量装配的说话人表征生成
作者：Yu-Huai Peng,Hung-Shin Lee,Pin-Tuan Huang,Hsin-Min Wang
机构*：Institute of Information Science, Academia Sinica, Taiwan
备注：Published in APSIPA ASC 2021
链接：点击下载PDF文件

【29】 FALCON: Fast Visual Concept Learning by Integrating Images, Linguistic descriptions, and Conceptual Relations
标题：Falcon：整合图像、语言描述和概念关系的快速视觉概念学习
作者：Lingjie Mei,Jiayuan Mao,Ziqi Wang,Chuang Gan,Joshua B. Tenenbaum
机构*：MIT CSAIL, UIUC, MIT-IBM Watson AI Lab, MIT BCS, CBMM, CSAIL
备注：First two authors contributed equally. Project page: this http URL
链接：点击下载PDF文件

【30】 Hybrid Handcrafted and Learnable Audio Representation for Analysis of Speech Under Cognitive and Physical Load
标题：用于认知和体力负荷下语音分析的手工和可学习混合音频表示
作者：Gasser Elbanna,Alice Biryukov,Neil Scheidwasser-Clow,Lara Orlandic,Pablo Mainar,Mikolaj Kegler,Pierre Beckmann,Milos Cernak
机构*： Ecole Polytechnique Federale de Lausanne (EPFL), Lausanne, Switzerland, Logitech Europe S.A., Lausanne, Switzerland, Imperial College London, London, United Kingdom, Universite de Lausanne, Lausanne, Switzerland
备注：Submitted to InterSpeech 2022
链接：点击下载PDF文件

【31】 Transformer Language Models without Positional Encodings Still Learn Positional Information
标题：没有位置编码的转换器语言模型仍然学习位置信息
作者：Adi Haviv,Ori Ram,Ofir Press,Peter Izsak,Omer Levy
机构*：τ Tel Aviv University, ω University of Washington, ι Intel Labs, Meta AI
链接：点击下载PDF文件

【32】 Knowledge-based Entity Prediction for Improved Machine Perception in Autonomous Systems
标题：基于知识的自主系统中改进机器感知的实体预测
作者：Ruwan Wickramarachchi,Cory Henson,Amit Sheth
机构*：Artificial Intelligence Institute, University of South Carolina, Columbia, SC, USA, Bosch Research and Technology Center, Pittsburgh, PA, USA
备注：6 pages, 4 figures
链接：点击下载PDF文件

【33】 Learning Local Displacements for Point Cloud Completion
标题：学习用于完成点云的局部位移
作者：Yida Wang,David Joseph Tan,Nassir Navab,Federico Tombari
机构*：Technische Universit¨at M¨unchen, Google Inc.
备注：Conference on Computer Vision and Pattern Recognition (CVPR) 2022
链接：点击下载PDF文件

【34】 Factored Adaptation for Non-Stationary Reinforcement Learning
标题：非平稳强化学习的因式自适应
作者：Fan Feng,Biwei Huang,Kun Zhang,Sara Magliacane
机构*： the environments and objectives are as- 1City University of Hong Kong 2Carnegie Mellon Univer-sity 3University of Amsterdam
备注：20 pages, 11 figures
链接：点击下载PDF文件

【35】 Mind the gap: Challenges of deep learning approaches to Theory of Mind
标题：心灵的鸿沟：深度学习方法对心理理论的挑战
作者：Jaan Aru,Aqeel Labash,Oriol Corcoll,Raul Vicente
机构*：Institute of Computer Science, University of Tartu
链接：点击下载PDF文件

【36】 Machine Learning Approaches for Non-Intrusive Home Absence Detection Based on Appliance Electrical Use
标题：基于家电使用的非侵入式家庭缺席检测的机器学习方法
作者：Athanasios Lentzas,Dimitris Vrakas
机构*：School of Informatics, Aristotle University of Thessaloniki, Thessaloniki, Greece, This research has been co-financed by the European Regional Development Fund of the, European Union and Greek national funds through the Operational Program Competitive-
备注：20 pages,submitted to "Expert Systems with Applications"
链接：点击下载PDF文件

【37】 Efficient Localness Transformer for Smart Sensor-Based Energy Disaggregation
标题：基于智能传感器的能量分解高效局部化转换器
作者：Zhenrui Yue,Huimin Zeng,Ziyi Kou,Lanyu Shang,Dong Wang
机构*：School of Information Sciences, University of Illinois Urbana-Champaign, Champaign, IL, USA
备注：Accepted to DCOSS 2022
链接：点击下载PDF文件

【38】 Recent improvements of ASR models in the face of adversarial attacks
标题：面对对抗性攻击的ASR模型的最新改进
作者：Raphael Olivier,Bhiksha Raj
机构*：Language Technologies Institute, Carnegie Mellon University, USA
备注：Submitted to Interspeech 2022
链接：点击下载PDF文件

【39】 Cross-modal Learning of Graph Representations using Radar Point Cloud for Long-Range Gesture Recognition
标题：用于远程手势识别的雷达点云图形表示的跨模式学习
作者：Souvik Hazra,Hao Feng,Gamze Naz Kiprit,Michael Stephan,Lorenzo Servadei,Rober Wille,Robert Weigel,Avik Santra
机构*：Infineon Technologies AG, Neubiberg, Germany, Friedrich-Alexander-University Erlangen-Nuremberg, Erlangen, Germany, Technical University of Munich, Munich, Germany, §equal contribution
备注：Submitted to IEEE Sensor Array and Multichannel Signal Processing Workshop (SAM 2022)
链接：点击下载PDF文件

【40】 An analytic theory for the dynamics of wide quantum neural networks
标题：广义量子神经网络动力学的解析理论
作者：Junyu Liu,Khadijeh Najafi,Kunal Sharma,Francesco Tacchino,Liang Jiang,Antonio Mezzacapo
机构*：Pritzker School of Molecular Engineering, The University of Chicago, Chicago, IL , USA, Chicago Quantum Exchange, Chicago, IL , USA, Kadanoff Center for Theoretical Physics, The University of Chicago, Chicago, IL , USA
备注：26 pages, 5 figures. Comments welcome
链接：点击下载PDF文件

【41】 Robust Disentangled Variational Speech Representation Learning for Zero-shot Voice Conversion
标题：用于零发声转换的稳健解缠变分语音表示学习
作者：Jiachen Lian,Chunlei Zhang,Dong Yu
机构*： UC Berkeley, EECS, CA , Tencent AI Lab, Bellevue, WA
备注：Accepted to 2022 ICASSP
链接：点击下载PDF文件

【42】 MAE-AST: Masked Autoencoding Audio Spectrogram Transformer
标题：MAE-AST：屏蔽自动编码音频频谱转换器
作者：Alan Baade,Puyuan Peng,David Harwath
机构*：Department of Computer Science, The University of Texas at Austin
备注：Submitted to INTERSPEECH. 5 pages, 2 figures, 5 tables
链接：点击下载PDF文件

【43】 Application of Ghost-DeblurGAN to Fiducial Marker Detection
标题：Ghost-DeblurGan在基准标记物检测中的应用
作者：Yibo Liu,Amaldev Haridevan,Hunter Schofield,Jinjun Shan
机构*： York University
备注：6 pages, 6 figures
链接：点击下载PDF文件
机器翻译，仅供参考

重大更新！公众号每日速递覆盖arXiv所有方向,涵盖CS|物理|数学|经济|统计|金融|生物|电气等领域。点击查看历史文章列表，可在公众号内订阅#arXiv每日学术速递话题，接收每日推送。点击访问arxivdaily.com每日更新更及时。