---

ACL2018読み会@名大

---

概要

• 会場: 名古屋大学工学部IB電子情報館中棟1階 IB011講義室 (MAP)
• 日時: 2018年9月10日(月) 10:10-17:15
• 紹介者(18名):
  • 武田浩一, 笹野遼平(名大武田・笹野研), 犬塚慎也, 古瀬弘樹, 土居裕典, Jung Dawoon(名大佐藤・松崎研),
    小川泰弘, 駒水孝裕, 山腰貴大, 藤岡和弥(名大外山研), 加藤芳秀(名大情報連携統括本部), 三輪誠, 辻村有輝(豊田工大),
    吉田光男, 廣中詩織(豊橋技科大), 綱川隆司(静大), 徳久良子, 小出智士
• 時間: 発表1件あたり15分、論文紹介12分前後、質疑応答・討論3分前後を想定

---

プログラム

• 10:10-10:15: Opening [笹野, slide]
• 10:15-11:20: Session 1: Information Extraction / Question Answering
  • Towards Understanding the Geometry of Knowledge Graph Embeddings [笹野, slide]
  • Improving Entity Linking by Modeling Latent Relations between Mentions [駒水, slide]
  • Semantically Equivalent Adversarial Rules for Debugging NLP models [三輪, slide]
  • Know What You Don’t Know: Unanswerable Questions for SQuAD [武田, slide]
• 12:30-13:35: Session 2: Summarization / Dialog / Discourse
  • Retrieve, Rerank and Rewrite: Soft Template Based Neural Summarization [小川, slide]
  • Unsupervised Abstractive Meeting Summarization with Multi-Sentence Compression and Budgeted Submodular Maximization [土居, slide]
  • Joint Reasoning for Temporal and Causal Relations [古瀬, slide]
  • Mem2Seq: Effectively Incorporating Knowledge Bases into End-to-End Task-Oriented Dialog Systems [徳久]
• 13:55-14:45: Session 3: Machine Translation / Generation
  • Context-Aware Neural Machine Translation Learns Anaphora Resolution [綱川, slide]
  • Document Context Neural Machine Translation with Memory Networks [山腰, slide]
  • Style Transfer Through Back-Translation [Jung, slide]
• 15:05-16:10: Session 4: Miscellaneous
  • SoPa: Bridging CNNs, RNNs, and Weighted Finite-State Machines [小出]
  • Transformation Networks for Target-Oriented Sentiment Classification [辻村, slide]
  • Straight to the Tree: Constituency Parsing with Neural Syntactic Distance [加藤]
  • Semi-supervised User Geolocation via Graph Convolutional Networks [廣中, slide]
• 16:25-17:15: Session 5: Semantics / Document Analysis
  • Sequence-to-Action: End-to-End Semantic Graph Generation for Semantic Parsing [犬塚, slide]
  • Joint Embedding of Words and Labels for Text Classification [藤岡, slide]
  • A Stylometric Inquiry into Hyperpartisan and Fake News [吉田]

---