---

ACL2016読み会@すずかけ台

---

概要

• 内容: ACL 2016で発表された論文を読む (+ BBQ)
• 場所: 東京工業大学すずかけ台キャンパスR2大会議室(6階)
• 日時: 2016年9月16日(金) 10:20-17:30

---

参加者

• 高村、笹野、上垣外、渡邉、石垣、瀧本、馬緤、菅原拓 (東工大)、
  小林、永井、菅原晃、佐野 (ヤフー)、横野 (富士通研)、萩行 (ウェザーニューズ) [=発表者:14名]
• 菊池、林、胡、佐藤貴、中村、村上、井田、金澤、田中、谷口、波多、矢田 (東工大)、田中 (サイバーエージェント)、
  小林、熊谷、樺山、濵園、松尾、大山、恒川 (小林研)、田中 (お茶大)、磯 (NAIST)、海野 (PFN) [=聴講者: 23名]

---

プログラム

• 10:20-10:30: Opening [笹野, pdf]
• 10:30-12:15: Session 1: Semantics / Word Embedding
  • Finding Non-Arbitrary Form-Meaning Systematicity Using String-Metric Learning for Kernel Regression [笹野, pdf]
  • Literal and Metaphorical Senses in Compositional Distributional Semantic Models [瀧本, pdf]
  • Intrinsic Subspace Evaluation of Word Embedding Representations [馬緤, pdf]
  • Diachronic Word Embeddings Reveal Statistical Laws of Semantic Change [高村, pdf]
• 13:15-14:20: Session 2: Machine Translation / Encoder-Decoder
  • Improving Neural Machine Translation Models with Monolingual Data [上垣外, pdf]
  • Models and Inference for Prefix-Constrained Machine Translation [小林]
  • Incorporating Copying Mechanism in Sequence-to-Sequence Learning [渡邉, pdf]
• 14:40-15:55: Session 3: Question Generation / Question Answering
  • Generating Factoid Questions With Recurrent Neural Networks: The 30M Factoid Question-Answer Corpus [佐野, pdf]
  • Generating Natural Questions About an Image [石垣, pdf]
  • A Parallel-Hierarchical Model for Machine Comprehension on Sparse Data [永井, pdf]
• 16:10-17:30: Session 4: Discourse / Neural Networks
  • Two Discourse Driven Language Models for Semantics [横野]
  • Chinese Zero Pronoun Resolution with Deep Neural Networks [萩行, pdf]
  • Globally Normalized Transition-Based Neural Networks [菅原晃, pdf]
  • Strategies for Training Large Vocabulary Neural Language Models [菅原拓]

---