论文
文章平均质量分 92
byn12345
好好学习,天天向上
展开
-
【论文解读 ACL 2020 | MixText】Linguistically-Informed Interpolation of Hidden Space for 半监督文本分类
论文题目:Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classification论文来源:ACL 2020论文链接:https://www.aclweb.org/anthology/2020.acl-main.194/代码链接:https://github.com/GT-SALT/MixText关键词:半监督;数据增强;文本分类;Mixup;BERT文章目录1 摘要2 引言2 相.原创 2021-02-26 17:49:21 · 1453 阅读 · 0 评论 -
【论文解读 AAAI 2021】Few-shot Learning for Multi-label Intent Detection
论文题目:Few-shot Learning for Multi-label Intent Detection论文来源:AAAI 2021论文链接:https://arxiv.org/abs/2010.05256代码链接:https://github.com/AtmaHou/FewShotMultiLabel关键词:少样本学习(FSL);多标签分类(MLC);用户意图检测;核回归;BERT;元学习文章目录1 摘要2 引言3 问题定义4 方法4.1 用于少样本多标签意图检测的框架4.2 元刻度阈.原创 2021-02-19 18:01:07 · 2069 阅读 · 0 评论 -
【论文解读 ASONAM 2019】Semi-Supervised Learning and Graph Neural Networks for Fake News Detection
论文题目:Semi-Supervised Learning and Graph Neural Networks for Fake News Detection论文链接:https://ieeexplore.ieee.org/document/9073625代码链接:无关键词:半监督学习;GNN;假新闻检测这篇文章就2页,简单记录一下。1 引言本文解决的是假新闻检测问题。作者提出基于图的半监督假新闻检测框架。作者认为基于图的模型表达能力较强,可以捕获文章间的上下文依赖,缓解标签不足的问题。.原创 2021-01-31 21:38:22 · 1140 阅读 · 0 评论 -
【论文解读 AAAI 2019 | UFD】Unsupervised Fake News Detection on Social Media: A Generative Approach
论文题目:Unsupervised Fake News Detection on Social Media: A Generative Approach论文来源:AAAI 2019论文链接:https://www.aaai.org/ojs/index.php/AAAI/article/view/4508代码链接:无关键词:无监督;假新闻检测;社交网络概率图推导看的眼花缭乱,这篇文章读得半懂不懂的,我数学太菜了…文章目录1 摘要2 引言3 模型3.1 分层用户参与3.2 问题建模3.3 概.原创 2021-01-31 13:03:06 · 1289 阅读 · 4 评论 -
【论文解读 COLING 2018 Best evaluation| SGM】SGM: Sequence Generation Model for Multi-Label Classification
论文题目:SGM: Sequence Generation Model for Multi-Label Classification论文来源:COLING 2018,Best evaluation论文链接:https://www.aclweb.org/anthology/C18-1330/代码链接:https://github.com/lancopku/SGM关键词:多标签分类(MLC),序列生成这篇文章是COLING 2018的Best evaluation文章目录1 摘要2 引言3 .原创 2021-01-27 21:14:05 · 483 阅读 · 0 评论 -
【论文解读 WSDM 2020 | KRF】Integrating Knowledge Relations into Style Correlations for 多标签音乐风格分类
论文题目:Beyond Statistical Relations: Integrating Knowledge Relations into Style Correlations for Multi-Label Music Style Classification论文来源:WSDM 2020论文链接:https://dl.acm.org/doi/10.1145/3336191.3371838代码链接:https://github.com/Makwen1995/MusicGenre关键词:音乐风格.原创 2021-01-26 22:40:53 · 532 阅读 · 0 评论 -
【论文解读 ICMR 2020 | KMGCN】Fake News Detection via Knowledge-driven Multimodal GCN
论文题目:Fake News Detection via Knowledge-driven Multimodal Graph Convolutional Networks论文来源:ICMR 2020论文链接:https://doi.org/10.1145/3372278.3390713关键词:假新闻检测;多模态;知识;图卷积文章目录1 摘要2 引言3 方法3.1 问题定义3.2 整体框架3.3 知识蒸馏3.4 多模态内容的图构建3.5 知识驱动的多模态图卷积网络(KMGCN)4 实验4.1 实验.原创 2020-07-23 23:12:29 · 4872 阅读 · 7 评论 -
【论文解读 MM 2019 | MKEMN】Multi-modal Knowledge-aware Event Memory Network for Social Media 谣言检测
论文题目:Multi-modal Knowledge-aware Event Memory Network for Social Media Rumor Detection论文来源:MM 2019论文链接:https://doi.org/10.1145/3343031.3350850关键词:多模态(文本+图像+知识);谣言检测;知识图谱;记忆网络(memory network);事件(event-level);注意力机制文章目录1 摘要2 引言3 相关工作3.1 谣言检测3.2 实体链接3.3 .原创 2020-07-19 00:42:06 · 3193 阅读 · 1 评论 -
【论文解读 WSDM 2019 | XBully】XBully: Cyberbullying Detection within a Multi-Modal Context
论文题目:XBully: Cyberbullying Detection within a Multi-Modal Context论文来源:WSDM 2019 刘欢老师组论文链接:https://doi.org/10.1145/3289600.3291037关键词:网络欺凌检测;多模态;异质图;网络嵌入文章目录1 摘要2 引言3 问题定义3.1 多模态网络欺凌检测3.2 通过多模态网络表示学习的网络欺凌检测3.3 挑战4 XBully模型4.1 Mode Hotspot Detection4.原创 2020-07-17 00:29:18 · 1166 阅读 · 0 评论 -
【论文解读 WWW 2019 | MVAE】Multimodal Variational Autoencoder for Fake News Detection
论文题目:MVAE: Multimodal Variational Autoencoder for Fake News Detection论文来源:WWW 2019论文链接:https://doi.org/10.1145/3308558.3313552代码链接:https://github.com/dhruvkhattar/MVAE关键词:多模态融合;图片;文本;变分自编码器;假新闻检测;microblogs文章目录1 摘要2 引言3 模型3.1 模型概览3.2 编码器3.3 解码器3.4 假.原创 2020-07-10 21:15:47 · 6734 阅读 · 4 评论 -
【论文解读 KDD 2018 | EANN】Event Adversarial Neural Networks for Multi-Modal Fake News Detection
论文题目:EANN: Event Adversarial Neural Networks for Multi-Modal Fake News Detection论文来源:KDD 2018论文链接:https://doi.org/10.1145/3219819.3219903代码链接:https://github.com/yaqingwang/EANN-KDD18关键词:事件;多模态(图像+文本);对抗神经网络;假新闻检测文章目录1 摘要2 模型3 实验4 总结1 摘要本文的目的是设计一个.原创 2020-07-10 20:30:27 · 2991 阅读 · 3 评论 -
【论文解读 EMNLP 2018】Cross-Lingual Cross-Platform Rumor Verification Pivoting on Multimedia Content
论文题目:Cross-Lingual Cross-Platform Rumor Verification Pivoting on Multimedia Content论文来源:EMNLP 2018论文链接:https://www.aclweb.org/anthology/D18-1385/代码链接:https://github.com/WeimingWen/CCRV关键词:跨语言;跨平台;特征;多媒体内容;谣言验证文章目录摘要1 引言2 相关工作3 CCMR 数据集4 框架概览5 跨语言跨平台.原创 2020-07-10 20:23:20 · 697 阅读 · 0 评论 -
【论文解读 ICDM 2019 | MVNN】Exploiting Multi-domain Visual Information for Fake News Detection
论文题目:Exploiting Multi-domain Visual Information for Fake News Detection论文来源:ICDM 2019论文链接:https://arxiv.org/abs/1908.04472关键词:假新闻检测;图像;频域;像素域;CNN;RNN;attention文章目录1 摘要2 引言3 模型3.1 模型概览3.2 频域子网络3.3 像素域子网络3.4 融合子网络4 实验5 总结1 摘要本文主要是利用新闻的图像信息,将频域和像素域的特.原创 2020-07-10 20:19:49 · 3098 阅读 · 3 评论 -
【论文解读 arXiv 2019 | DEAN】DEAN: Learning Dual Emotion for Fake News Detection on Social Media
论文题目:DEAN: Learning Dual Emotion for Fake News Detection on Social Media论文来源:arXiv 2019论文链接:https://arxiv.org/abs/1903.01728关键词:假新闻检测,社交网络,情感,multimodal,Gate文章目录1 摘要2 引言3 模型3.1 建模发布者情感3.2 建模社交情感3.3 DEAN框架4 实验5 总结1 摘要本文要解决的任务是利用文本中的情感信息,进行假新闻检测。在许.原创 2020-07-10 19:50:49 · 1453 阅读 · 1 评论 -
【论文解读 MM 2017 | att-RNN】Multimodal Fusion with RNNs for Rumor Detection on Microblogs
论文题目:Multimodal Fusion with Recurrent Neural Networks for Rumor Detection on Microblogs论文来源:MM 2017论文链接:https://doi.org/10.1145/3123266.3123454关键词:多模态融合,谣言检测,LSTM,注意力机制,microblog文章目录1 摘要2 引言3 模型3.1 模型概览3.2 文本和社交上下文的联合表示3.3 图像的视觉表示3.4 用于视觉表示的注意力3.5 模型.原创 2020-07-10 19:32:57 · 5160 阅读 · 11 评论 -
fake news相关 2019-2020 五篇论文阅读
创新点、改进点、实验用到的数据集、不足文章目录1 Bi-GCN数据集本文的亮点和要点思考2 Capturing the Style of Fake News数据集已有方法的问题本文的亮点和要点思考3 WeFEND数据集已有方法的不足文章的亮点和要点思考4 Proactive Discovery of Fake News Domains from Real-Time Social Media Feeds数据集文章的亮点和要点思考5 dEFEND数据集本文的亮点和要点思考References1 Bi.原创 2020-07-02 21:58:04 · 3993 阅读 · 1 评论 -
【论文解读 AAAI 2020 | Bi-GCN】Rumor Detection on Social Media with Bi-Directional GCN
论文题目:Rumor Detection on Social Media with Bi-Directional Graph Convolutional Networks论文来源:AAAI 2020 清华大学, 腾讯AI论文链接:https://arxiv.org/abs/2001.06362代码链接:https://github.com/TianBian95/BiGCN关键词:谣言检测,社交媒体,GCN文章目录1 摘要2 引言3 Preliminaries4 Bi-GCN谣言检测模型5 实验.原创 2020-06-11 23:47:56 · 12898 阅读 · 14 评论 -
【论文翻译 假新闻检测综述 HICSS 2019】Can Machines Learn to Detect Fake News? A Survey Focused on Social Media
论文题目:Can Machines Learn to Detect Fake News? A Survey Focused on Social Media论文来源:HICSS 2019,Proceedings of the 52nd Hawaii International Conference on System Sciences论文链接:https://www.researchgate.net/publication/330364905_Can_Machines_Learn_To_Detect_F.原创 2020-06-10 22:18:21 · 1996 阅读 · 1 评论 -
【论文翻译 虚假新闻检测综述 KDD 2017】Fake News Detection on Social Media: A Data Mining Perspective
论文题目:Fake News Detection on Social Media: A Data Mining Perspective论文来源:KDD 2017 美国亚利桑那州立大学, 美国密歇根州立大学论文链接:https://arxiv.org/abs/1708.01967关键词:虚假新闻检测,社交媒体,综述文章目录1 摘要2 引言3 假新闻的特点3.1 虚假新闻的定义3.2 传统新闻媒体上的虚假新闻3.3 社交媒体上的虚假新闻4 假新闻检测4.1 问题定义4.2 特征抽取4.2.1 新闻内.原创 2020-06-09 23:35:17 · 5387 阅读 · 0 评论 -
【转载】沈向洋:读论文的三个层次
本文转自AI科技评论:沈向洋:读论文的三个层次作者 | 蒋宝尚修改| 沈向洋编辑 | 丛末沈向洋:“要养成写小总结的习惯,最好能做报告,这样真的能够增加你对所读论文的理解。”“或许你永远不知道你以前读过的书能在什么时候能够派上用场,但是请保持阅读,因为阅读的过程也是在你大脑中建立认知的过程。”目录1快速阅读:划分结构层次2仔细阅读:批判思维3创造性阅读:积极思考5月14日,沈向洋博士在全球创新学院(GIX)课程上曾做了一场线上公开课《You ar...转载 2020-06-02 12:24:01 · 650 阅读 · 0 评论 -
【文献解读 情感合成】Expressive Speech Driven Talking Avatar Synthesis with DBLSTM using 有限的情感双峰数据
论文题目:Expressive Speech Driven Talking Avatar Synthesis with DBLSTM using Limited Amount of Emotional Bimodal Data论文来源:INTERSPEECH 2016 清华大学, 香港中文大学论文链接:https://isca-speech.org/archive/Interspeech_2016/pdfs/0364.PDF关键词:合成会讲话的虚拟人,情感,嘴唇动作,面部表情,DBLSTM这.原创 2020-05-27 22:20:44 · 561 阅读 · 0 评论 -
【论文解读 ICLR 2020】Differentiable Learning of Numerical Rules in Knowledge Graphs
论文题目:Differentiable learning of numerical rules in knowledge graphs论文来源:ICLR 2020 CMU, Bosch Center for AI论文链接:https://openreview.net/forum?id=rJleKgrKwS关键词:知识图谱,数值规则,规则的学习(抽取),知识推理文章目录1 摘要2 引言3 相关工作4 一些定义与准备工作5 使用数值特征和否定词进行规则学习5.1 比较操作符6 实验7 总结参考文.原创 2020-05-20 12:26:06 · 1973 阅读 · 0 评论 -
【论文解读 AAAI2020|AliNet】KnowledgeGraph Alignment Network with Gated Multi-hop Neighborhood Aggregation
论文题目:Knowledge Graph Alignment Network with Gated Multi-hop Neighborhood Aggregation论文来源:AAAI 2020 南京大学,阿里巴巴论文链接:https://arxiv.org/abs/1911.08936代码链接:关键词:知识图谱,实体对齐,GNN文章目录1 摘要2 引言3 准备工作3.1 GNN3.2 知识图谱的实体对齐3.3 用于实体对齐的GNN4 知识图谱对齐网络4.1 Gated Multi-hop.原创 2020-05-16 15:47:58 · 4822 阅读 · 4 评论 -
【论文翻译 AAAI 2020 | 知识图谱综述】A Survey on Knowledge Graphs: Representation, Acquisition and Applications
论文题目:A Survey on Knowledge Graphs: Representation, Acquisition and Applications论文来源:AAAI 2020 Philip S. Yu等人关键词:知识图谱,表示学习,知识图谱补全,关系抽取,知识推理待更新…PS. 以后再也不翻译综述了,实在是太长了,也看不下去[捂脸哭泣]。个人感觉这篇文章看一看每个小结的总结就行了,中间主要是方法的列举。(都翻译了这么多,也不想接下来只翻译小结,还是善始善终吧)文章目录1.原创 2020-05-14 21:20:48 · 17142 阅读 · 14 评论 -
【论文解读 AAAI 2020 | GNTP】Differentiable Reasoning on Large Knowledge Bases and Natural Language
论文题目:Differentiable Reasoning on Large Knowledge Bases and Natural Language论文来源:AAAI 2020 伦敦大学, Facebook论文链接:https://arxiv.org/abs/1912.10824代码链接:https://github.com/uclnlp/gntp关键词:可微推理,机器推理,知识库,自然语言文本,attention文章目录1 摘要2 引言3 端到端的可微性证明4 在大规模KBs上有效的可.原创 2020-05-09 12:35:33 · 1544 阅读 · 0 评论 -
【论文解读 ICLR 2020 | Jure Leskovec组】Query2box: Reasoning over KGs in Vector Space using Box Embedding
论文题目:Query2box: Reasoning over Knowledge Graphs in Vector Space using Box Embeddings论文来源:ICLR 2020 斯坦福大学 Jure Leskovec组论文链接:https://arxiv.org/abs/2002.05969代码链接:http://snap.stanford.edu/query2bo...原创 2020-05-05 21:54:24 · 4853 阅读 · 0 评论 -
【论文解读 AAAI 2020】Graph-Based Reasoning over Heterogeneous External Knowledge for 常识问答
论文题目:Graph-Based Reasoning over Heterogeneous External Knowledge for Commonsense Question Answering论文来源:AAAI 2020 信工所,北大,微软论文链接:https://arxiv.org/abs/1909.05311关键词:知识图谱,机器推理,常识问答,GCN,Attention...原创 2020-04-29 22:25:09 · 4291 阅读 · 2 评论 -
【转载】图神经网络的新基准Benchmarking Graph Neural Networks
本文转自知乎文章:图神经网络的新基准Benchmarking Graph Neural Networks最近GNN备受关注,相信大家也都能感受到。但是,一旦我们开始阅读相关论文,开展相关的实验时,会发现一些问题。我们一般会从节点分类数据集cora, citeseer, pubmed,图分类PROTEINS, NCI1, NCI109等数据集入手,这些数据集相对都比较小,数据集小不是什么问题...转载 2020-04-29 11:15:37 · 590 阅读 · 1 评论 -
【论文解读 KDD 2018 | PinSage】Graph Convolutional Neural Networks for Web-Scale Recommender Systems
论文链接:https://arxiv.org/abs/1806.01973论文来源:KDD 2018 斯坦福大学, Pinterest关键词:GCN,推荐系统,web-scale,PinSage本文在GraphSAGE基础上进行改进(去掉了需要将整张图存储到内存中的限制),提出PinSage模型,成功在Pinterest的推荐系统上得到了应用,第一个将图卷积(GCN)在超大规模数据...原创 2020-04-27 12:19:09 · 4899 阅读 · 0 评论 -
【论文解读IJCAI 2019】Extracting Entities and Events as a Single Task Using a Transition-Based NeuralModel
论文题目:Extracting Entities and Events as a Single Task Using a Transition-Based Neural Model论文来源:IJCAI 2019 武汉大学, 东华大学, 西湖大学论文链接:https://www.ijcai.org/Proceedings/2019/753代码链接:https://github.com/z...原创 2020-04-24 23:01:00 · 1031 阅读 · 0 评论 -
【论文解读 WWW 2019 | FBMA】Event Detection using Hierarchical Multi-Aspect Attention
论文题目:Event Detection using Hierarchical Multi-Aspect Attention论文来源:WWW 2019 乔治梅森大学,弗吉尼亚理工大学论文链接:https://doi.org/10.1145/3308558.3313659代码链接:https://github.com/sumehta/FBMA关键词:分层注意力,多角度注意力,事件检测,...原创 2020-04-23 23:59:11 · 1352 阅读 · 0 评论 -
【论文解读 EMNLP 2019】Joint Event and Temporal RE with Shared Representations and Structured Prediction
论文题目:Joint Event and Temporal Relation Extraction with Shared Representations and Structured Prediction论文来源:EMNLP 2019 南加利福尼亚大学,伊利诺伊大学香槟分校论文链接:https://www.aclweb.org/anthology/D19-1041/关键词:联合学习,...原创 2020-04-22 23:50:49 · 910 阅读 · 1 评论 -
【论文解读 NAACL 2019 | DyGIE】A General Framework for Information Extraction using Dynamic Span Graphs
论文题目:A General Framework for Information Extraction using Dynamic Span Graphs论文来源:NAACL 2019 Google AI Language, 华盛顿大学论文链接:https://www.aclweb.org/anthology/N19-1308/代码链接:https://github.com/luany...原创 2020-04-21 23:59:41 · 6436 阅读 · 0 评论 -
【论文解读 EMNLP 2019 | DyGIE++】Entity, Relation, and EE with Contextualized Span Representations
论文题目:Entity, Relation, and Event Extraction with Contextualized Span Representations论文来源:EMNLP 2019 华盛顿大学, Google AI Language论文链接:https://www.aclweb.org/anthology/D19-1585/代码链接:https://github.c...原创 2020-04-21 14:10:16 · 4707 阅读 · 0 评论 -
【论文解读 EMNLP 2019 | HMEAE】Hierarchical Modular Event Argument Extraction
论文题目:HMEAE: Hierarchical Modular Event Argument Extraction论文来源:EMNLP 2019 清华、微信AI论文链接:https://www.aclweb.org/anthology/D19-1584/代码链接:https://github.com/thunlp/HMEAE关键词:事件元素抽取(EAE),概念层次,attention...原创 2020-04-20 17:44:19 · 2690 阅读 · 2 评论 -
【论文解读 EMNLP 2019 | TLNN】Event Detection with Trigger-Aware Lattice Neural Network
论文题目:Event Detection with Trigger-Aware Lattice Neural Network论文来源:EMNLP 2019 清华论文链接:https://www.aclweb.org/anthology/D19-1033/代码链接:https://github.com/thunlp/TLNN关键词:事件检测,word-trigger不匹配,trigger...原创 2020-04-19 18:59:50 · 1879 阅读 · 3 评论 -
【论文解读 ACL 2019 | PLMEE】Exploring Pre-trained Language Models for Event Extraction and Generation
论文题目:Exploring Pre-trained Language Models for Event Extraction and Generation论文来源:ACL 2019 国防科技大学论文链接:https://www.aclweb.org/anthology/P19-1522/关键词:事件抽取,预训练语言模型,事件生成,角色重叠,训练数据不足文章目录1 摘要2 引言3 ...原创 2020-04-15 12:49:42 · 6065 阅读 · 13 评论 -
【论文解读 EMNLP 2019 | MOGANED】ED with Multi-Order Graph Convolution and Aggregated Attention
论文题目:Event Detection with Multi-Order Graph Convolution and Aggregated Attention论文来源:EMNLP 2019论文链接:https://www.aclweb.org/anthology/D19-1582/代码链接(非官方):https://github.com/wzq016/MOGANED-Implementa...原创 2020-04-11 14:42:45 · 2214 阅读 · 1 评论 -
【论文解读 ICLR 2020 | DropEdge】TOWARDS DEEP GRAPH CONVOLU-TIONAL NETWORKS ON NODE CLASSIFICATION
论文题目:DROPEDGE: TOWARDS DEEP GRAPH CONVOLU-TIONAL NETWORKS ON NODE CLASSIFICATION论文来源:ICLR 2020论文链接:https://openreview.net/forum?id=Hkx1qkrKPr代码链接:https://github.com/DropEdge/DropEdge关键词:GNN的加深,过拟...原创 2020-04-10 23:54:37 · 8028 阅读 · 15 评论 -
【论文解读 EMNLP 2018 | JMEE】Jointly Multiple EE via Attention-based Graph Information Aggregation
论文题目:Jointly Multiple Events Extraction via Attention-based Graph Information Aggregation论文来源:EMNLP 2018论文链接:https://arxiv.org/abs/1809.09078代码链接:https://github.com/lx865712528/EMNLP2018-JMEE关键词:...原创 2020-04-09 15:36:27 · 4186 阅读 · 1 评论