出品 | 深度學習這件小事公眾號
自然語言處理（11月11日更新版）
[1] DoLFIn: Distributions over Latent Features for Interpretability作者 | Phong Le, Willem Zuidema連結 | https://arxiv.org/abs/2011.05295 [2] Neural Machine Translation for Extremely Low-Resource African Languages: A Case Study on Bambara作者 | Allahsera Auguste Tapo, Bakary Coulibaly, Sébastien Diarra, Christopher Homan, Julia Kreutzer, Sarah Luger, Arthur Nagashima, Marcos Zampieri, Michael Leventhal連結 | https://arxiv.org/abs/2011.05284 [3] Towards Interpretable Natural Language Understanding with Explanations as Latent Variables作者 | Wangchunshu Zhou, Jinyi Hu, Hanlin Zhang, Xiaodan Liang, Maosong Sun, Chenyan Xiong, Jian Tang連結 | https://arxiv.org/abs/2011.05268 [4] Medical Knowledge-enriched Textual Entailment Framework作者 | Shweta Yadav, Vishal Pallagani, Amit Sheth連結 | https://arxiv.org/abs/2011.05257 [5] Towards Preemptive Detection of Depression and Anxiety in Twitter作者 | David Owen, Jose Camacho Collados, Luis Espinosa-Anke連結 | https://arxiv.org/abs/2011.05249 備註 | Social Media Mining for Health Applications (#SMM4H) | COLING 2020[6] On the State of Social Media Data for Mental Health Research作者 | Keith Harrigian, Carlos Aguirre, Mark Dredze連結 | https://arxiv.org/abs/2011.05233 備註 | Originally submitted to ICWSM in January 2020. Updated November 2020. [7] UmBERTo-MTSA @ AcCompl-It: Improving Complexity and Acceptability Prediction with Multi-task Learning on Self-Supervised Annotations連結 | https://arxiv.org/abs/2011.05197 備註 | 5 pages, Best system award for the AcCompl-It shared task at the EVALITA 2020 workshop[8] Multi-Task Sequence Prediction For Tunisian Arabizi Multi-Level Annotation作者 | Elisa Gugliotta (1,2,3), Marco Dinarelli (2), Olivier Kraif (3) ((1) Sapienza University of Rome, (2) Université Grenoble Alpes - Laboratoire LIG, (Getalp group), (3) Université Grenoble Alpes- Laboratoire LIDILEM)連結 | https://arxiv.org/abs/2011.05152 備註 | Paper accepted at the Fifth Arabic Natural Language Processing Workshop (WANLP) 2020[9] Does Social Support Expressed in Post Titles Elicit Comments in Online Substance Use Recovery Forums?作者 | Anietie Andy, Sharath Guntuku連結 | https://arxiv.org/abs/2011.05103 [10] Translating Similar Languages: Role of Mutual Intelligibility in Multilingual Transformers作者 | Ife Adebara, El Moatez Billah Nagoudi, Muhammad Abdul Mageed連結 | https://arxiv.org/abs/2011.05037 [11] To What Degree Can Language Borders Be Blurred In BERT-based Multilingual Spoken Language Understanding?作者 | Quynh Do, Judith Gaspers, Tobias Roding, Melanie Bradford連結 | https://arxiv.org/abs/2011.05007 [12] When Do You Need Billions of Words of Pretraining Data?作者 | Yian Zhang, Alex Warstadt, Haau-Sing Li, Samuel R. Bowman連結 | https://arxiv.org/abs/2011.04946 [13] On the Usefulness of Self-Attention for Automatic Speech Recognition with Transformers作者 | Shucong Zhang, Erfan Loweimi, Peter Bell, Steve Renals連結 | https://arxiv.org/abs/2011.04906 [14] Determining Question-Answer Plausibility in Crowdsourced Datasets Using Multi-Task Learning作者 | Rachel Gardner, Maya Varma, Clare Zhu, Ranjay Krishna連結 | https://arxiv.org/abs/2011.04883 備註 | Published at the 6th Workshop on Noisy User-generated Text (W-NUT) 2020 at EMNLP (6 pages, 4 figures)[15] A Transfer Learning Approach for Dialogue Act Classification of GitHub Issue Comments作者 | Ayesha Enayet, Gita Sukthankar連結 | https://arxiv.org/abs/2011.04867 備註 | Poster at 2020 International Conference on Social Informatics[16] Natural Language Inference in Context -- Investigating Contextual Reasoning over Long Texts作者 | Hanmeng Liu, Leyang Cui, Jian Liu, Yue Zhang連結 | https://arxiv.org/abs/2011.04864 [17] Simultaneous Speech-to-Speech Translation System with Neural Incremental ASR, MT, and TTS作者 | Katsuhito Sudoh, Takatomo Kano, Sashi Novitasari, Tomoya Yanagita, Sakriani Sakti, Satoshi Nakamura連結 | https://arxiv.org/abs/2011.04845 [18] Multi-document Summarization via Deep Learning Techniques: A Survey作者 | Congbo Ma, Wei Emma Zhang, Mingyu Guo, Hu Wang, Quan Z. Sheng連結 | https://arxiv.org/abs/2011.04843 [19] Language Through a Prism: A Spectral Approach for Multiscale Language Representations作者 | Alex Tamkin, Dan Jurafsky, Noah Goodman連結 | https://arxiv.org/abs/2011.04823 [20] EstBERT: A Pretrained Language-Specific BERT for Estonian作者 | Hasan Tanvir, Claudia Kittask, Kairit Sirts連結 | https://arxiv.org/abs/2011.04784 [21] An Analysis of Dataset Overlap on Winograd-Style Tasks作者 | Ali Emami, Adam Trischler, Kaheer Suleman, Jackie Chi Kit Cheung連結 | https://arxiv.org/abs/2011.04767 備註 | 11 pages with references, accepted at COLING 2020[22] Adversarial Semantic Collisions作者 | Congzheng Song, Alexander M. Rush, Vitaly Shmatikov連結 | https://arxiv.org/abs/2011.04743 [23] CLAR: A Cross-Lingual Argument Regularizer for Semantic Role Labeling作者 | Ishan Jindal, Yunyao Li, Siddhartha Brahma, Huaiyu Zhu連結 | https://arxiv.org/abs/2011.04732 備註 | EMNLP 2020, ACL Findings[24] Biomedical Information Extraction for Disease Gene Prioritization作者 | Jupinder Parmar, William Koehler, Martin Bringmann, Katharina Sophia Volz, Berk Kapicioglu連結 | https://arxiv.org/abs/2011.05188 備註 | Knowledge Representation and Reasoning Meets Machine Learning Workshop(KR2ML), at NeurIPS 2020[25] Generalized LSTM-based End-to-End Text-Independent Speaker Verification作者 | Soroosh Tayebi Arasteh連結 | https://arxiv.org/abs/2011.04896 備註 | 7 pages, 7 tables, 6 figures. Research Internship at Fraunhofer Institute for Integrated Circuits IIS, in cooperation with Pattern Recognition Lab at Friedrich-Alexander-University Erlangen-Nuremberg, Germany. Re-implementation of the paper arXiv:1710.10467 by Wan et al[26] Pretraining Strategies, Waveform Model Choice, and Acoustic Configurations for Multi-Speaker End-to-End Speech Synthesis作者 | Erica Cooper, Xin Wang, Yi Zhao, Yusuke Yasuda, Junichi Yamagishi連結 | https://arxiv.org/abs/2011.04839 [27] Personalized Query Rewriting in Conversational AI Agents作者 | Alireza Roshan-Ghias, Clint Solomon Mathialagan, Pragaash Ponnusamy, Lambert Mathias, Chenlei Guo連結 | https://arxiv.org/abs/2011.04748 [28] Speaker De-identification System using Autoencodersand Adversarial Training作者 | Fernando M. Espinoza-Cuadros, Juan M. Perero-Codosero, Javier Antón-Martín, Luis A. Hernández-Gómez連結 | https://arxiv.org/abs/2011.04696