TOHOKU NLP LAB - 東北大学 乾・鈴木研究室TOHOKU NLP LAB - 東北大学 乾・鈴木研究室

MENU

Natural Language Processing

Explore Human & Machine Thinking

News【ニュース】

2019 Aug.15
News Release

Acceptance to EMNLP-IJCNLP 2019 / EMNLP-IJCNLP 2019 採択

The following papers have been accepted to the 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP).

EMNLP-IJCNLP 2019に下記の論文が採択されました。

  • Hiroki Ouchi, Jun Suzuki and Kentaro Inui.
    “Transductive Learning of Neural Language Models for Syntactic and Semantic Analysis”
  • Shun Kiyono, Jun Suzuki, Masato Mita, Tomoya Mizumoto and Kentaro Inui.
    “An Empirical Study of Incorporating Pseudo Data into Grammatical Error Correction”
  • Xiaoyu Shen, Jun Suzuki, Kentaro Inui, Hui Su, Dietrich Klakow and Satoshi Sekine.
    “Select and Attend: Towards Controllable Content Selection in Text Generation”
  • Masato Hagiwara, Takumi Ito, Tatsuki Kuribayashi, Jun Suzuki and Kentaro Inui.
    “TEASPN: Framework and Protocol for Integrated Writing Assistance Environments” (Demo Track)
open
2019 Aug.02
News Release

The 57th Annual Meeting of the Association for Computational Linguistics (ACL2019)

The following works have been presented at the main conference and workshops of the 57th Annual Meeting of the Association for Computational Linguistics (ACL2019).

2019年7月28日から8月2日にかけて開催された the 57th Annual Meeting of the Association for Computational Linguistics (ACL2019) および併催ワークショップで下記の発表を行いました。

Main Conference:

  • Tatsuki Kuribayashi, Hiroki Ouchi, Naoya Inoue, Paul Reisert, Toshinori Miyoshi, Jun Suzuki and Kentaro Inui.
    “An Empirical Study of Span Representations in Argumentation Structure Parsing”
  • Benjamin Heinzerling and Michael Strube.
    “Sequence Tagging with Contextual and Non-Contextual Subword Representations: A Multilingual Evaluation”
  • Motoki Sato, Jun Suzuki and Shun Kiyono.
    “Effective Adversarial Regularization for Neural Machine Translation Models”

Workshops:


open
2019 Aug.02
Talk

MiCS: Dr. Kevin Duh / MiCS (みちのく情報伝達学セミナー) : Kevin Duh氏ご講演

Dr. Kevin Duh from John Hopkins University visited the lab and gave a lecture titled “Membership Inference Attacks on Sequence-to-Sequence Models”. (MiCS)

Johns Hopkins UniversityのKevin Duh氏にお越しいただき、「Membership Inference Attacks on Sequence-to-Sequence Models」という表題でご講演いただきました。(みちのく情報伝達学セミナー)

open
2019 Jul.31
News Release

Open Campus held / オープンキャンパスが開催されました

An Open Campus event targeting prospective students and the local community has been held on July 30th and 31st. The Inui-Suzuki Lab showcased their research results with demonstration systems and panel displays.

7月30日-31日の2日間、入学希望者や地域の方々を対象にしたオープンキャンパスが開催されました。 乾・鈴木研究室では、研究成果のデモンストレーションやパネル展示をおこないました。

After the two-day event, Lab members enjoyed a social gathering.

また、二日間のオープンキャンパス後には研究室メンバー皆で打ち上げも行いました。

open
2019 Jul.12
News Release

Third International Workshop on Symbolic-Neural Learning (SNL-2019)

Associate Professor Jun Suzuki gave an invited talk at the Third International Workshop on Symbolic-Neural Learning (SNL-2019) held from July 11 to July 12, 2019 on the following topic:

  • “Usability Enhancements to Neural Word Embeddings”

2019年7月11日から7月12日にかけて開催された Third International Workshop on Symbolic-Neural Learning (SNL-2019) にて、鈴木潤准教授が下記の題目で招待講演をおこないました。

  • “Usability Enhancements to Neural Word Embeddings”

open
2019 Jul.11
News Release

Selection at Google AI for Japan / Google AI for Japan 採択

Associate Professor Jun Suzuki has been selected for “Google AI for Japan” under the following research project proposal:

  • Kentaro Inui “Language Model as a Knowledge Base: Toward Integration of Knowledge Representation, Acquisition, Retrieval, and Inference”

“Google AI for Japan is a program that supports the development of the next generation of AI human resources, promotes AI utilization for solving business and social issues, and aims to contribute to AI research in Japan.” (Google AI for Japan: AI人材の育成と技術活用促進を目指して)

乾教授が下記の研究課題で「Google AI for Japan」に採択されました。「『Google AI for Japan』は、次世代の AI 人材育成を支援するとともに、ビジネスや社会課題解決に向けた AI 活用促進、さらに国内の AI 研究への貢献を目的」とした助成プログラムです。(Google AI for Japan: AI人材の育成と技術活用促進を目指して

  • 乾 健太郎 「知識ベースとしての言語モデル:知識の表現・獲得・検索・推論の統合に向けて」
open
2019 Jul.09
News Release

Writing assistance environment protocol and framework TEASPN released / ライティング支援環境のためのプロトコルおよびフレームワークTEASPNの公開

The TEASPN (Text Editing Assistance Smartness Protocol for Natural Language) protocol and framework for writing assistance environments has been released. It is a standardized mechanism for linking natural language processing modules such as error correction and auto-completion with editors.
The project’s main contributors are Dr. Masato Hagiwara (Octanove Labs/RIKEN AIP) and 2nd year Master students Takumi Ito and Tatsuki Kuribayashi (Langsmith Inc.).

TEASPN — Framework and Protocol for Integrated Writing Assistance Environments – youtube

ライティング支援環境を実現するプロトコルおよびフレームワークTEASPN (Text Editing Assistance Smartness Protocol for Natural Language) を公開しました。誤り訂正・自動補完などの自然言語処理モジュールとエディタとの橋渡しをする仕組みを標準化したものとなっています。
Masato Hagiwara (Octanove Labs/RIKEN AIP) さん、伊藤拓海、栗林樹生(修士2年/Langsmith Inc.)が中心となって取り組んだプロジェクトです。

TEASPN — Framework and Protocol for Integrated Writing Assistance Environments – youtube

open
2019 Jun.28
Talk

MiCS: Dr. Patrycja Świeczkowska / MiCS (みちのく情報伝達学セミナー) : Patrycja Świeczkowska氏ご講演

Dr. Patrycja Świeczkowska from the Hokkaido University visited the Lab and gave a presentation titled “Towards a motivational dialogue system”. MiCS

北海道大学のPatrycja Świeczkowska氏にお越しいただき、「Towards a motivational dialogue system」という表題でご講演いただきました。(みちのく情報伝達学セミナー)

open
2019 Jun.26
News Release

2019 1st Information Processing Society Seminar “Developing a Data Driven New Era” / 情報処理学会 連続セミナー2019「データ駆動で新時代を切り拓く」

Associate Professor Jun Suzuki gave an invited talk at the 2019 1st Information Processing Society Seminar “Developing a Data Driven New Era”: The Future with AI (1): The Latest Trends in Natural Language Processing (Japanese only) under the following subject:

  • “Interpretability and explainability of natural language processing with deep learning”

Additionally, he appeared in a panel discussion at the seminar.

情報処理学会 連続セミナー2019「データ駆動で新時代を切り拓く」第1回:AIと歩む未来(1):自然言語処理の最新動向 にて、鈴木潤准教授が下記の題目で招待講演をおこないました。

  • 「深層学習による自然言語処理の解釈性/説明性」

また、同セミナーのパネル討論に出演しました。

open
2019 Jun.19
News Release

Information Processing Society Tohoku Branch Student Encouragement Award / 情報処理学会東北支部学生奨励賞受賞

1st year Master student Kotaro Kitayama recieved the Information Processing Society Tohoku Branch Student Encouragement Award. Congratulations!

北山晃太郎(修士1年)が情報処理学会東北支部学生奨励賞を受賞しました。おめでとうございます!

open
2019 Jun.19
Blog

Group photo update / 集合写真を撮りました

We updated our group photo on the members page. Let’s have another great year together!

メンバーページの集合写真を更新しました。これからもどうぞよろしくお願いいたします。

open
2019 Jun.07
News Release

2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics(NAACL-HLT 2019)

The following works have been presented at the 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics(NAACL-HLT 2019) main conference and workshops, held from June 2nd to June 7th.

2019年6月2日から6月7日にかけて開催された2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics(NAACL-HLT 2019)および併催WSにおいて、下記の発表を行いました。

Main Conference:

  • Shota Sasaki, Jun Suzuki and Kentaro Inui
    “Subword-based Compact Reconstruction of Word Embeddings.”
  • Masato Mita, Tomoya Mizumoto, Masahiro Kaneko, Ryo Nagata and Kentaro Inui
    “Cross-Corpora Evaluation and Analysis of Grammatical Error Correction Models — Is Single-Corpus Evaluation Enough?”

Workshops:

open
2019 Jun.07
News Release

Presented at JSAI 2019 / 2019年度 第33回 人工知能学会全国大会(JSAI)

The following works have been presented at 33rd National Meeting of The Japanese Society for Artificial Intelligence (JSAI 2019) held from June 4th to June 7th 2019.

  • Kaori Abe, Jun Suzuki, Masaaki Nagata, Kentaro Inui
    “Construction of Coherent Translation Evaluation Dataset in Machine Translation”
  • Taku Nakamura, Ran Tian, Shota Sasaki, Kentaro Inui
    “Distinguish Antonyms in Multiple Viewpoints on Distributed Word Representations”

2019年6月4日から6月7日にかけて開催された2019年度 第33回 人工知能学会全国大会(JSAI2019) において、下記の発表を行いました。

  • 阿部 香央莉, 鈴木 潤, 永田 昌明, 乾 健太郎
    “機械翻訳における訳語一貫性評価用データセットの構築”
  • 中村 拓, 田 然, 佐々木 翔大, 乾 健太郎
    “単語埋め込みにおける複数視点の対義語判定”

open
2019 May.28
Talk

MiCS: visit by Mr. Kim Geewook / MiCS (みちのく情報伝達学セミナー) : Geewook Kim氏ご講演

Mr. Kim Geewook from Kyoto University visited the lab, and gave a presentation on “Representation Learning with Weighted Inner Product for Universal Approximation of General Similarities”. MiCS

京都大学のGeewook Kim氏にお越しいただき、「重み付き内積による一般類似度関数の万能近似性と表現学習への応用」という表題でご講演いただきました。(みちのく情報伝達学セミナー)

open
2019 May.27
News Release

Acceptance to ACL SRW / ACL SRW 採択

The following papers have been accepted to the 2019 ACL Student Research Workshop (SRW).
The 2019 ACL Student Research Workshop (SRW)に下記の論文が採択されました。

  • Diana Galván-Sosa.
    “Active Reading Comprehension: A dataset for learning the Question-Answer Relationship strategy”
  • Farjana Sultana Mim, Naoya Inoue, Paul Reisert, Hiroki Ouchi and Kentaro Inui.
    “Unsupervised Learning of Discourse-Aware Text Representation for Essay Scoring”
open
2019 May.27
News Release

Acceptance to RepL4NLP / RepL4NLP 採択

The following paper has been accepted to the 4th Workshop on Representation Learning for NLP (RepL4NLP).
The 4th Workshop on Representation Learning for NLP (RepL4NLP)に下記の論文が採択されました。

  • Federico Lopez, Benjamin Heinzerling and Michael Strube.
    “Fine-grained Entity Typing in Hyperbolic Space”
open
2019 May.27
News Release

Acceptance to BEA / BEA 採択

The following paper has been accepted to the 14th Workshop on Innovative Use of NLP for Building Educational Applications (BEA).
The 14th Workshop on Innovative Use of NLP for Building Educational Applications (BEA)に下記の論文が採択されました。

  • Tomoya Mizumoto, Hiroki Ouchi, Yoriko Isobe, Paul Reisert, Ryo Nagata, Satoshi Sekine and Kentaro Inui.
    “Analytic Score Prediction and Justification Identification in Automated Short Answer Scoring”
  • Hiroki Asano, Masato Mita, Tomoya Mizumoto, Jun Suzuki.
    “The AIP-Tohoku System at the BEA-2019 Shared Task”
open
2019 May.21
News Release

Acceptance to BlackboxNLP / BlackboxNLP 採択

The following paper has been accepted to BlackboxNLP workshop.

BlackboxNLP workshop に下記の論文が採択されました。

  • Hitomi Yanaka, Koji Mineshima, Daisuke Bekki, Kentaro Inui, Satoshi Sekine, Lasha Abzianidze and Johan Bos.
    “Can neural networks understand monotonicity reasoning?”
open
2019 May.15
News Release

Acceptance to ACL 2019 / ACL 2019 採択

The following papers have been accepted to the 57th Annual Meeting of the Association for Computational Linguistics (ACL).

The 57th Annual Meeting of the Association for Computational Linguistics (ACL)に下記の論文が採択されました。

  • Tatsuki Kuribayashi, Hiroki Ouchi, Naoya Inoue, Paul Reisert, Toshinori Miyoshi, Jun Suzuki and Kentaro Inui.
    “An Empirical Study of Span Representations in Argumentation Structure Parsing”
  • Benjamin Heinzerling and Michael Strube.
    “Sequence Tagging with Contextual and Non-Contextual Subword Representations: A Multilingual Evaluation”
  • Motoki Sato, Jun Suzuki and Shun Kiyono.
    “Effective Adversarial Regularization for Neural Machine Translation Models”
open
2019 May.13
Talk

MiCS: Talk by Mr. Yuto Yoneyama, Mr. Yuta Tomomatsu, and Mr. Takamichi Toda from CyberAgent/ MiCS (みちのく情報伝達学セミナー) : 米山結人氏、友松祐太氏、戸田隆道氏ご講演

Mr. Yuto Yoneyama, Mr. Yuta Tomomatsu, and Mr. Takamichi Toda from CyberAgent, inc. gave a lecture on “The real issues facing chatbot products and the future”. (MiCS)

サイバーエージェントの皆さまにお越しいただき、米山結人氏、友松祐太氏、戸田隆道氏に「チャットボットプロダクトが抱えるリアルな課題とこれから」という表題でご講演いただきました。(みちのく情報伝達学セミナー)

open
View More News

これ以上は記事がありません

これ以上は記事がありません

Recent Changes
Close