丸山 拓海

長岡技術科学大学大学院 電気電子情報工学専攻 修士2年
自然言語処理研究室

Mail: maruyama@jnlp.org

■ B3勉強会
    自然言語処理のための機械学習
     第1回 文書及び単語の数学的表現 [slide]
       第2回 凝集型クラスタリング / k-means [slide]
      第3回 混合正規分布を用いたクラスタリング / EM algorithm [slide]
      第4回 ナイーブベイズ分類器_1 (多変数ベルヌーイモデル) [slide]
      第5回 ナイーブベイズ分類器_2 (多変数ベルヌーイモデル) [slide]
      第6回 ナイーブベイズ分類器_3 (多項モデル) [slide]
      第7回 サポートベクトルマシン [slide]
      第8回 カーネル法 / 対数線形モデル [slide]
      第9回 系列ラベリング  [slide]

■ 文献紹介
    2017年3月 〜
  1. 新聞記事中の難解語を平易な表現へ変換する手法の提案 [paper] [slide]
  2. 言語横断質問応答に適した機械翻訳評価尺度の調査  [paper] [slide]
  3. Simple English Wikipedia: A New Text Simplification Task [paper] [slide]
  4. Lexical Simplification with Neural Ranking [paper] [slide]
  5. OptimizingStatistical Machine Translation for Text Simplification [paper] [slide]
  6. Automatic Text Simplification for Spanish: Comparative Evaluation of Various Simplification Strategies [paper] [slide]
  7. Translating from Original to Simplified sentences using Moses: When does it Actually work? [paper] [slide]
  8. Exploring Neural Text Simplification Models [paper] [slide]
  9. Sentence Alignment Methods for Improving Text Simplification Systems [paper] [slide]

    2018年3月 〜
  1. Split and Rephrase [paper] [slide]
  2. Supervised Learning of Universal Sentence Representations from Natural Language Inference Data [paper] [slide]
  3. Sentence Simplification with Memory-Augmented Neural Networks [slide]
  4. Sentence Simplification with Deep Reinforcement Learning [paper][slide]
  5. Simplification Using Paraphrases and Context-based Lexical Substitution [paper][slide]
  6. An Operation Network for Abstractive Sentence Compression [paper][slide]
  7. Retrieve, Rerank and Rewrite: Soft Template Based Neural Summarization [paper][slide]
  8. BLEU is Not Suitable for the Evaluation of Text Simplification [paper][slide]
  9. Don’t Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization [paper][slide]
  10. Integrating Transformer and Paraphrase Rules for Sentence Simplification [paper][slide]
  11.  Understanding Back-Translation at Scale [paper][slide] 
  12.  Back-Translation Sampling by Targeting Difficult Words in Neural Machine Translation [paper][slide]

    2019年3月 〜
  1. Dynamic Data Selection for Neural Machine Translation [paper][slide]
  2. Extract and Edit: An Alternative to Back-Translation for Unsupervised Neural Machine Translation [paper][slide]
  3. Simple Unsupervised Keyphrase Extraction using Sentence Embeddings [paper][slide]
  4.  Addressing Troublesome Words in Neural Machine Translation [paper][slide]
  5. An Embarrassingly Simple Approach for Transfer Learning from Pretrained Language Models [paper][slide]
  6. Soft Contextual Data Augmentation for Neural Machine Translation [paper][slide]
  7. Hint-Based Training for Non-Autoregressive Machine Translation [paper][slide]
  8. What makes a good conversation? How controllable attributes affect human judgments [paper][slide]
  9. Controlling Text Complexity in Neural Machine Translation [paper][slide]
  10.  Simple Unsupervised Summarization by Contextual Matching [paper][slide]
  11.  Misspelling Oblivious Word Embeddings [paper][slide]
  12.  Variational Pretraining for Semi-supervised Text Classification [paper][slide]
■ 参加報告
       言語処理学会第24回年次大会 [slide]
       The 32nd Pacific Asia Conference on Language, Information and Computation (PACLIC 32)

       言語処理学会第25回年次大会 [slide]

■ 論文
  1. ⼭本 和英, 丸⼭ 拓海, ⾓張 ⻯晴, 稲岡 夢⼈, ⼩川 耀⼀朗, 勝⽥ 哲弘, 髙橋 寛治. やさしい日本語対訳コーパスの構築. 言語処理学会第23回年次大会, pp.763-766 (2017.3)
  2. Takumi Maruyama and Kazuhide Yamamoto. Sentence Simplification with Core Vocabulary. Proceedings of the International Conference on Asian Language Processing (IALP 2017), pp.363-366 (2017.12) [Paper] [Slide] [Abstract]
  3. Takumi Maruyama and Kazuhide Yamamoto. Simplified Corpus with Core Vocabulary. The 11th International Conference on Language Resources and Evaluation (LREC 2018), pp.1153-1160 (2018.5) [Paper] [Abstract] [Poster] [Page]
  4. Takumi Maruyama and Kazuhide Yamamoto. Lexical Substitution is Practical for Rare Word Simplification. The 32nd Pacific Asia Conference on Language,  Information and Computation (PACLIC 32)