自然语言处理每日论文速递[12.29]

访问arxivdaily.com，获取带摘要的学术速递，更多学科、收藏、评论、搜索……应有尽有。同步公众号(arXiv每日学术速递)，欢迎关注

cs.CL 方向，今日共计36篇

【1】 Universal Sentence Representation Learning with Conditional Masked Language Model
标题：基于条件掩蔽语言模型的通用句子表征学习
作者：Ziyi Yang,Yinfei Yang,Daniel Cer,Jax Law,Eric Darve
机构*： Stanford University, Google Al
备注：preprint
链接：https://arxiv.org/abs/2012.14388

【2】 DeepHateExplainer: Explainable Hate Speech Detection in Under-resourced Bengali Language
标题：DeepHateExplainer：资源不足的孟加拉语中可解释的仇恨语音检测
作者：Md. Rezaul Karim,Sumon Kanti Dey,Bharathi Raja Chakravarthi
机构*：rezaul.karim(fit. fraunhofer.de, Fraunhofer FIT, Noakhali Science and Technology, Insight Centre for Data Analytics, RWTH Aachen University, National University of Ireland, Aachen, Germany, Noakhali, Bangladesh, Galway, Ireland
备注：Extended version of this paper is currently under review in the IEEE Access journal
链接：https://arxiv.org/abs/2012.14353

【3】 BURT: BERT-inspired Universal Representation from Learning Meaningful Segment
标题：BURT：从学习有意义的片段中获得BERT启发的通用表征
作者：Yian Li,Hai Zhao
备注：arXiv admin note: text overlap with arXiv:2009.04656
链接：https://arxiv.org/abs/2012.14320

【4】 Panarchy: ripples of a boundary concept
标题：泛政体：边界概念的涟漪
作者：Juan Rocha,Linda Luvuno,Jesse Rieb,Erin Crockett,Katja Malmborg,Michael Schoon,Garry Peterson
机构*：Stockholm Resilience Centre, Stockholm University, Kraftriket ,B, Stockholm, Future Earth, The Swedish Royal Academy of Sciences, Lilla Frescativagen ,A, Stockholm, South American Institute for Resilience and Sustainability Studies, Maldonado, Uruguay, Center for Complex Systems in Transition Stellenbosch University, South Africa, McGill University, rue Sherbrooke ouest, Montreal (Quebec) H,A ,B, Canada, McGill University, Lakeshore Road, Sainte-, Anne-de-Bellevue, QC, H,X ,V, Canada, Arizona State University, Tempe, Arizona, USA
备注：11 pages, 5 figures
链接：https://arxiv.org/abs/2012.14312

【5】 Towards Fully Automated Manga Translation
标题：迈向全自动漫画翻译
作者：Ryota Hinami,Shonosuke Ishiwatari,Kazuhiko Yasuda,Yusuke Matsui
机构*： Mantra Inc., Yahoo Japan Corporation, The University of Tokyo
备注：Accepted to AAAI 2021
链接：https://arxiv.org/abs/2012.14271

【6】 Red Dragon AI at TextGraphs 2020 Shared Task: LIT : LSTM-Interleaved Transformer for Multi-Hop Explanation Ranking
标题：TextGraphs 2020共享任务上的Red Dragon AI：LIT：LSTM-用于多跳解释排名的交错转换器
作者：Yew Ken Chia,Sam Witteveen,Martin Andrews
机构*：Red Dragon Al, Singapore
备注：Accepted paper for TextGraphs-14 workshop at COLING 2020. (6 pages including references)
链接：https://arxiv.org/abs/2012.14164

【7】 On Generating Extended Summaries of Long Documents
标题：关于长篇文献的扩展摘要生成问题
作者：Sajad Sotudeh,Arman Cohan,Nazli Goharian
机构*：十, IR Lab, Georgetown University, Washington D.C., USA, Allen Institute for Artificial, Seattle, WA, USA
备注：Accepted at SDU 2021
链接：https://arxiv.org/abs/2012.14136

【8】 Neural Text Generation with Artificial Negative Examples
标题：基于人工反例的神经文本生成
作者：Keisuke Shirai,Kazuma Hashimoto,Akiko Eriguchi,Takashi Ninomiya,Shinsuke Mori
机构*：Salesforce Research Microsoft Research, Ehime University, Academic Center for Computing and Media Studies, Kyoto University
链接：https://arxiv.org/abs/2012.14124

【9】 Syntax-Enhanced Pre-trained Model
标题：语法增强型预训练模型
作者：Zenan Xu,Daya Guo,Duyu Tang,Qinliang Su,Linjun Shou,Ming Gong,Wanjun Zhong,Xiaojun Quan,Nan Duan,Daxin Jiang
机构*：Sun Yat-sen University, Microsoft,Beijing, China
链接：https://arxiv.org/abs/2012.14116

【10】 Pivot Through English: Reliably Answering Multilingual Questions without Document Retrieval
标题：透视英文：无需文档检索即可可靠地回答多语言问题
作者：Ivan Montero,Shayne Longpre,Ni Lao,Andrew J. Frank,Christopher DuBois
机构*：University of Washington, Apple Inc., Christopher Dubois
链接：https://arxiv.org/abs/2012.14094

【11】 Automatic Curriculum Learning With Over-repetition Penalty for Dialogue Policy Learning
标题：对话策略学习中带有过度重复惩罚的自动课程学习
作者：Yangyang Zhao,Zhenyu Wang,Zhenhua Huang
机构*：South China University of Technology
链接：https://arxiv.org/abs/2012.14072

【12】 ALP-KD: Attention-Based Layer Projection for Knowledge Distillation
标题：ALP-KD：基于注意力的知识提炼层次投影
作者：Peyman Passban,Yimeng Wu,Mehdi Rezagholizadeh,Qun Liu
机构*：Huawei Noah's Ark Lab
备注：AAAI 2021. This work has been done while Peyman Passban was at Huawei
链接：https://arxiv.org/abs/2012.14022

【13】 SMART: A Situation Model for Algebra Story Problems via Attributed Grammar
标题：SMART：基于属性文法的代数故事题情境模型
作者：Yining Hong,Qing Li,Ran Gong,Daniel Ciao,Siyuan Huang,Song-Chun Zhu
机构*：University of California, Los Angeles, USA.
备注：None
链接：https://arxiv.org/abs/2012.14011

【14】 Explaining NLP Models via Minimal Contrastive Editing (MiCE)
标题：通过最小对比编辑解释NLP模型(MICE)
作者：Alexis Ross,Ana Marasović,Matthew E. Peters
机构*：Allen Institute for Artificial Intelligence, Seattle, WA, USA, University of Washington
链接：https://arxiv.org/abs/2012.13985

【15】 MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding Pretraining
标题：MEDIE：自然语言理解预训练医学缩略语消歧数据集
作者：Zhi Wen,Xing Han Lu,Siva Reddy
机构*：McGill University, Facebook CIFAR AI Chair, Mila-Quebec Artificial Intelligence Institute
备注：EMNLP 2020 Clinical NLP
链接：https://arxiv.org/abs/2012.13978

【16】 Adaptive Convolution for Semantic Role Labeling
标题：用于语义角色标注的自适应卷积
作者：Kashif Munir,Hai Zhao,Zuchao Li
备注：This submission is an early version of our recently accepted IEEE/ACM TASLP paper
链接：https://arxiv.org/abs/2012.13939

【17】 SG-Net: Syntax Guided Transformer for Language Representation
标题：SG-Net：语法制导的语言表示转换器
作者：Zhuosheng Zhang,Yuwei Wu,Junru Zhou,Sufeng Duan,Hai Zhao,Rui Wang
备注：The early version accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). arXiv admin note: text overlap with arXiv:1908.05147
链接：https://arxiv.org/abs/2012.13915

【18】 An Embarrassingly Simple Model for Dialogue Relation Extraction
标题：一种简单得令人尴尬的对话关系抽取模型
作者：Fuzhao Xue,Aixin Sun,Hao Zhang,Eng Siong Chng
机构*： Institute of High Performance Computing, ASTAR, Singapore
链接：https://arxiv.org/abs/2012.13873

【19】 My Teacher Thinks The World Is Flat! Interpreting Automatic Essay Scoring Mechanism
标题：我的老师认为世界是平的！解读作文自动评分机制
作者：Swapnil Parekh,Yaman Kumar Singla,Changyou Chen,Junyi Jessy Li,Rajiv Ratn Shah
机构*： IIIT-Delhi, Adobe, ykumarQadobe. com, State University of New York at Buffalo, University of Texas at Austin
链接：https://arxiv.org/abs/2012.13872

【20】 Learning Light-Weight Translation Models from Deep Transformer
标题：从深层Transformer学习轻量化平移模型
作者：Bei Li,Ziyang Wang,Hui Liu,Quan Du,Tong Xiao,Chunliang Zhang,Jingbo Zhu
机构*：INLP Lab, Northeastern University, Shenyang, China, NiuTrans Research, Shenyang, China
备注：Accepted by AAAI2021
链接：https://arxiv.org/abs/2012.13866

【21】 Inserting Information Bottlenecks for Attribution in Transformers
标题：在Transformer中插入属性的信息瓶颈
作者：Zhiying Jiang,Raphael Tang,Ji Xin,Jimmy Lin
机构*：University of Waterloo
备注：Accepted by EMNLP2020 Findings
链接：https://arxiv.org/abs/2012.13838

【22】 Fine-grained Emotion and Intent Learning in Movie Dialogues
标题：电影对话中的细粒度情感和意图学习
作者：Anuradha Welivita,Yubo Xie,Pearl Pu
机构*：Ecole polytechnique federale de Lausanne, Switzerland
备注：8 pages
链接：https://arxiv.org/abs/2012.13624

【23】 LOREN: Logic Enhanced Neural Reasoning for Fact Verification
标题：Loren：用于事实验证的逻辑增强型神经推理
作者：Jiangjie Chen,Qiaoben Bao,Jiaze Chen,Changzhi Sun,Hao Zhou,Yanghua Xiao,Lei Li
机构*：Shanghai Key Laboratory of Data Science, Fudan University, China, ByteDance Lab, Beijing, China
链接：https://arxiv.org/abs/2012.13577

【24】 Contextual Temperature for Language Modeling
标题：语言建模的语境温度
作者：Pei-Hsin Wang,Sheng-Iou Hsieh,Shih-Chieh Chang,Yu-Ting Chen,Jia-Yu Pan,Wei Wei,Da-Chang Juan
机构*： National Tsing Hua University, Hsinchu, Taiwan , Google, Mountain View, CA, USA
链接：https://arxiv.org/abs/2012.13575

【25】 Towards a Universal Continuous Knowledge Base
标题：走向通用的连续知识库
作者：Gang Chen,Maosong Sun,Yang Liu
机构*：Institute for AI Industry Research, Tsinghua University, Institute for Artificial Intelligence, Tsinghua University, Beijing National Research Center for Information Science and Technology, Beijing Academy of Artificial Intelligence
链接：https://arxiv.org/abs/2012.13568

【26】 Why Neural Machine Translation Prefers Empty Outputs
标题：为什么神经机器翻译更喜欢空输出
作者：Xing Shi,Yijun Xiao,Kevin Knight
机构*：DiDi Labs, Admiralty Way, UC Santa Barbara, Marina del Rey, Marina del rey
备注：6 pages
链接：https://arxiv.org/abs/2012.13454

【27】 ThamizhiUDp: A Dependency Parser for Tamil
标题：ThamizhiUDp：泰米尔语依存关系解析器
作者：Kengatharaiyer Sarveswaran,Gihan Dias
机构*：University of Moratuwa Sri Lanka University of Moratuwa Sri Lanka
备注：5 Pages, Published at ICON2020: 17th International Conference on Natural Language Processing (December 18-21, 2020)
链接：https://arxiv.org/abs/2012.13436

【28】 Mechanism of Evolution Shared by Gene and Language
标题：基因与语言共同的进化机制
作者：Li-Min Wang,Hsing-Yi Lai,Sun-Ting Tsai,Shan-Jyun Wu,Meng-Xue Tsai,Daw-Wei Wang,Yi-Ching Su,Chen Siang Ng,Tzay-Ming Hong
机构*：National Tsing Hua University, Hsinchu , Taiwan, Republic of China, (Dated: December ,)
备注：15 pages, 13 figures, 3 table
链接：https://arxiv.org/abs/2012.14309

【29】 The Curse of Dense Low-Dimensional Information Retrieval for Large Index Sizes
标题：大索引密集低维信息检索的诅咒
作者：Nils Reimers,Iryna Gurevych
机构*：Ubiquitous Knowledge Processing Lab(UKP-TUDA), Technical University of Darmstadt, www.ukp. tu-darmstadt. de
链接：https://arxiv.org/abs/2012.14210

【30】 Neural document expansion for ad-hoc information retrieval
标题：面向即席信息检索的神经文档扩展
作者：Cheng Tang,Andrew Arnold
机构*：Amazon
链接：https://arxiv.org/abs/2012.14005

【31】 Measuring University Impact: Wikipedia approach
标题：衡量大学影响力：维基百科方法
作者：Tatiana Kozitsina,Viacheslav Goiko,Roman Palkin,Valentin Khomutenko,Yulia Mundrievskaya,Maria Sukhareva,Isak Froumin,Mikhail Myagkov
机构*：IMoscow Institute of Physics and Technology (State University), Applied Mathematics, Institutskiy per., Dolgoprudny, Moscow Region, Russian, Federal Research Center Computer Science and Control, Russian Academy of Sciences, Tomsk State University, Laboratory of Big Data in Social Sciences, Lenin Ave., Tomsk, Moscow State University, Public administration faculty,(,) Lomonosov Pr., Moscow, SInstitute of Education, Potpovskiy Pereulok, building , Moscow, Russian Federation., University of Oregon, Oregon, United States., Corresponding authore-mail: http://babkinatsyandex.ru)
链接：https://arxiv.org/abs/2012.13980

【32】 Improving Opinion Spam Detection by Cumulative Relative Frequency Distribution
标题：利用累积相对频率分布改进评论垃圾邮件检测
作者：Michela Fazzolari,Francesco Buccafurri,Gianluca Lax,Marinella Petrocchi
备注：Manuscript accepted for publication in ACM Journal of Data and Information Quality. This is the pre-final version, before proofs checking
链接：https://arxiv.org/abs/2012.13905

【33】 Translating Natural Language Instructions to Computer Programs for Robot Manipulation
标题：将自然语言指令翻译成用于机器人操作的计算机程序
作者：Sagar Gubbi Venkatesh,Raviteja Upadrashta,Bharadwaj Amrutur
备注：Under review for ICRA 2021
链接：https://arxiv.org/abs/2012.13695

【34】 Spatial Reasoning from Natural Language Instructions for Robot Manipulation
标题：基于自然语言指令的机器人操作空间推理
作者：Sagar Gubbi Venkatesh,Anirban Biswas,Raviteja Upadrashta,Vikram Srinivasan,Partha Talukdar,Bharadwaj Amrutur
备注：Under review for ICRA 2021
链接：https://arxiv.org/abs/2012.13693

【35】 Social media data reveals signal for public consumer perceptions
标题：社交媒体数据揭示了公众消费者感知的信号
作者：Neeti Pokhriyal,Abenezer Dara,Benjamin Valentino,Soroush Vosoughi
机构*：Dartmouth College
备注：In Proceedings of the ACM International Conference on AI in Finance (ICAIF '20)
链接：https://arxiv.org/abs/2012.13675

【36】 Learning by Fixing: Solving Math Word Problems with Weak Supervision
标题：在固定中学习：在弱监督下解决数学应用题
作者：Yining Hong,Qing Li,Daniel Ciao,Siyuan Haung,Song-Chun Zhu
机构*：University of California, Los Angeles, USA., Problem: A truck travels , kilometers in , hours. At this
备注：None
链接：https://arxiv.org/abs/2012.10582

机器翻译，仅供参考；“机构”仅供参考