ITPub博客

首页 > 人工智能 > > ACL 2019最佳论文出炉

ACL 2019最佳论文出炉

作者:兔子爱喝奶茶 时间:2019-08-01 10:49:14 0 删除 编辑

ACL 2019最佳论文出炉:华人团队包揽最佳长短论文一作,中科院、华为等榜上有名

自然语言处理领域的顶会ACL 2019公布最佳论文,本次共颁布了4个奖项,共有8个获奖名额,分别是:

  1. 最佳长论文
  2. 最佳短论文
  3. 最佳DEMO论文
  4. 5篇杰出论文奖

今年,华人一作团队拿下了最佳长论文、最佳短论文和2篇杰出论文,中科院、中国科学院大学、腾讯、华为诺亚方舟实验室、南京理工大学、香港理工大学等榜上有名。今年的ACL 2019空前热闹。据ACL官方发布的数据显示,今年共收到2906篇投稿,相比于去年的1544篇投稿量增加了75%以上。


最佳长论文获奖者是中国的研究人员,这篇论文名为:

《Bridging the Gap between Training and Inference for Neural Machine Translation》

论文地址:https://arxiv.org/abs/1906.02448


最佳短论文:

《Do you know that Florence is packed with visitors?Evaluating state-of-the-art models of speaker commitment》(最佳短论文的一作也是华人,俄亥俄州立大学博士生蒋南江,二作则是该校助理教授Marie-Catherine de Marneffe。)

论文地址:https://linguistics.osu.edu/people/jiang.1879


最佳Demo论文:

《OpenKiwi: An Open Source Framework for Quality Estimation》

今年的最佳Demo论文授予了Unbabel团队,他们提出了一个基于Pytorch的开源框架OpenKiwi,用于评估神经机器翻译质量。

论文地址:https://arxiv.org/abs/1902.08646

代码地址:https://github.com/Unbabel/OpenKiwi


5篇杰出论文:

1、Emotion-Cause Pair Extraction: A New Task to Emotion Analysis in Texts

https://arxiv.org/abs/1906.01267

作者:Rui Xia, Zixiang Ding(南京理工大学)

2、A Simple Theoretical Model of Importance for Summarization

https://www.aclweb.org/anthology/P19-1101

作者:Maxime Peyrard(EPFL)

3、Transferable Multi-Domain State Generator for Task-Oriented Dialogue Systems

https://arxiv.org/abs/1905.08743

作者:Chien-Sheng Wu, Andrea Madotto, Ehsan Hosseini-Asl, Caiming Xiong, Richard Socher and Pascale Fung(香港科技大学、Salesforce等)

4、We need to talk about standard splits

https://wellformedness.com/papers/gorman-bedrick-2019.pdf

作者:Kyle Gorman and Steven Bedrick (纽约城市大学、俄勒冈健康与科学大学等)

5、Zero-Shot Entity Linking by Reading Entity Descriptions

https://arxiv.org/abs/1906.07348

作者:Lajanugen Logeswaran, Ming-Wei Chang, Kenton Lee, Kristina Toutanova, Jacob Devlin and Honglak Lee(密歇根大学、谷歌等)


40%提名论文一作为华人

从论文作者角度分析,在这32篇提名论文中,13篇论文的第一作者是华人,占比40%。

这13篇华人一作的论文分别是:

1、Detecting Concealed Information in Text and Speech

作者:Shengli Hu(康奈尔大学)

https://www.aclweb.org/anthology/P19-1039

2、AMR Parsing as Sequence-to-Graph Transduction

作者:Sheng Zhang(约翰霍普金斯大学)等

https://arxiv.org/abs/1905.08704

3、Transferable Multi-Domain State Generator for Task-Oriented

作者:Chien-Sheng Wu(港科大), Andrea Madotto, Ehsan Hosseini-Asl等

https://arxiv.org/abs/1905.08743

4、A Modularized, Versatile, and Extensible Toolkit for Text Generation

作者:Zhiting Hu胡志挺(CMU), Haoran Shi, Bowen Tan等

https://www.aclweb.org/anthology/W18-2503

5、Emotion-Cause Pair Extraction: A New Task to Emotion Analysis in Texts

作者:Rui Xia(南京理工),Zixiang Ding

https://arxiv.org/abs/1906.01267

6、Visually Grounded Neural Syntax Acquisition

作者:Haoyue Shi(丰田工业大学芝加哥分校), Jiayuan Mao, Kevin Gimpel and Karen Livescu

https://arxiv.org/abs/1906.02890

7、An Imitation Learning Approach to Unsupervised Parsing

作者:Bowen Li(爱丁堡大学), Lili Mou, Frank Keller

https://arxiv.org/abs/1906.02276

8、Decomposable Neural Paraphrase Generation

作者:Zichao Li(华为诺亚方舟实验室), Xin Jiang, Lifeng Shang and Qun Liu

https://arxiv.org/abs/1906.09741

9、Robust Neural Machine Translation with Doubly Adversarial Inputs

作者:Yong Cheng(Google AI), Lu Jiang and Wolfgang Macherey

https://arxiv.org/abs/1906.02443

10、Bridging the Gap between Training and Inference for Neural Machine Translation

作者:Wen Zhang(中科院), Yang Feng, Fandong Meng, Di You and Qun Liu

https://arxiv.org/abs/1906.02448

11、Do you know that Florence is packed with visitors? Evaluating state-of-the-art models of speaker commitment

作者:Nanjiang Jiang(the ohio state uiversity)等

12、ConvLab: Multi-Domain End-to-End Dialog System Platform

作者:Sungjin Lee(微软研究院), Qi Zhu, Ryuichi Takanobu等

https://arxiv.org/abs/1904.08637

13、Persuasion for Good: Towards a Personalized Persuasive Dialogue System for Social Good

作者:Xuewei Wang(浙大), Weiyan Shi等

https://arxiv.org/abs/1906.06725


传送门

ACL提名论文名单:

http://www.acl2019.org/EN/nominations-for-acl-2019-best-paper-awards.xhtml

(来源:量子位)

来自 “ ITPUB博客 ” ,链接:https://blog.itpub.net/69918097/viewspace-3138580/ ,如需转载,请注明出处,否则将追究法律责任。

请登录后发表评论 登录
全部评论

注册时间:2019-04-26

  • 博文量
    1470
  • 访问量
    11177