Hayate Iso

Research Scientist at Megagon Labs.
hayate@megagon.ai

Curriculum Vitae

Hellođź‘‹ I am a research scientist at Megagon Labs. I am interested in natural language generation, opinion summarization, and user-generated text analysis.

Our team is looking for research student interns. Please feel free to contact me if you're interested.


Publications

2022

Comparative Opinion Summarization via Collaborative Decoding
Hayate Iso, Xiaolan Wang, Stefanos Angelidis, Yoshihiko Suhara
Findings of the Association for Computational Linguistics (ACL), 2022

Paper arXiv Github
BibTeX


@inproceedings{iso2022comparative,
  title = {Comparative Opinion Summarization via Collaborative Decoding},
  author = {Hayate Iso and Xiaolan Wang and Stefanos Angelidis and Yoshihiko Suhara},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2022},
  url = {https://aclanthology.org/2022.findings-acl.261/},
  year = {2022},
}

2021

Convex Aggregation for Opinion Summarization
Hayate Iso, Xiaolan Wang, Yoshihiko Suhara, Stefanos Angelidis, Wang-Chiew Tan
Findings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 2021

Paper arXiv Github
BibTeX


@inproceedings{iso21emnlpfindings,
  title = {{C}onvex {A}ggregation for {O}pinion {S}ummarization},
  author = {Hayate Iso and Xiaolan Wang and Yoshihiko Suhara and Stefanos Angelidis and Wang{-}Chiew Tan},
  booktitle = {Findings of the Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  month = {November},
  year = {2021}
  }

End-to-end Biomedical Entity Linking with Span-based Dictionary Matching
Shogo Ujiie, Hayate Iso, Shuntaro Yada, Shoko Wakamiya, Eiji Aramaki
Biomedical Natural Language Processing Workshop (BioNLP), 2021

Paper arXiv
BibTeX


@inproceedings{ujiie2021end2end,
  title = {End-to-end Biomedical Entity Linking with Span-based Dictionary Matching},
  author = {Shogo Ujiie and Hayate Iso and Shuntaro Yada and Shoko Wakamiya and Eiji Aramaki},
  booktitle = {Proceedings of the 20th Workshop on Biomedical Language Processing},
  address = {Online},
  month = {June},
  pages = {162--167},
  url = {https://aclanthology.org/2021.bionlp-1.18},
  year = {2021},
}

Biomedical Entity Linking with Contrastive Context Matching
Shogo Ujiie, Hayate Iso, Eiji Aramaki
preprint

arXiv Github
BibTeX


@article{ujiie2021context,
  title = {Biomedical Entity Linking with Contrastive Context Matching},
  author = {Shogo Ujiie and Hayate Iso and Eiji Aramaki},
  url = {https://arxiv.org/abs/2106.07583},
  year = {2021},
}

2020

Fact-based Text Editing
Hayate Iso, Chao Qiao, Hang Li
Association for Computational Linguistics (ACL), 2020
Acceptance rate: 778/3,429 = 22.7 %

Paper arXiv Github Slide
BibTeX


@inproceedings{iso2020fact,
  title = {{F}act-based {T}ext {E}diting},
  author = {Hayate Iso and Chao Qiao and Hang Li},
  booktitle = {Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL)},
  address = {Online},
  month = {July},
  pages = {171--182},
  url = {https://www.aclweb.org/anthology/2020.acl-main.17},
  year = {2020},
}

2019

Learning to Select, Track, and Generate for Data-to-Text
Hayate Iso, Yui Uehara, Tatsuya Ishigaki, Hiroshi Noji, Eiji Aramaki, Ichiro Kobayashi, Yusuke Miyao, Naoaki Okazaki and Hiroya Takamura
Association for Computational Linguistics (ACL), 2019
Acceptance rate: 660/2,905 = 22.7 %

Paper arXiv Github Poster
BibTeX


@inproceedings{iso2019learning,
  title = {{L}earning to {S}elect, {T}rack, and {G}enerate for {D}ata-to-{T}ext},
  author = {Hayate Iso and Yui Uehara and Tatsuya Ishigaki and Hiroshi Noji and Eiji Aramaki and Ichiro Kobayashi and Yusuke Miyao and Naoaki Okazaki and Hiroya Takamura},
  booktitle = {Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL)},
  address = {Florence, Italy},
  month = {August},
  pages = {2102--2113},
  url = {https://www.aclweb.org/anthology/P19-1202},
  year = {2019},
}

2018

Wisdom in Adversity: A Twitter Study of the Japanese Tsunami
Paolo Casani, Hayate Iso, Shoko Wakamiya, Eiji Aramaki
Advances in Social Networks Analysis and Mining (ASONAM), 2018

Paper PDF
BibTeX


@inproceedings{paolo2018wisdom,
  title = {Wisdom in Adversity: A Twitter Study of the Japanese Tsunami.},
  author = {Casani, Paolo and Iso, Hayate and Wakamiya, Shoko and Aramaki, Eiji},
  booktitle = {Advances in Social Networks Analysis and Mining (ASONAM)},
  address = {Barcelona, Spain},
  month = {August},
  pages = {750--754},
  url = {https://ieeexplore.ieee.org/abstract/document/8508253},
  year = {2018},
}

Parsing Japanese Tweets into Universal Dependencies
Hayate Iso, Kaoru Ito, Hiroyuki Nagai, Taro Okahisa and Eiji Aramaki
Universal Dependencies Workshop (UDW), 2018

Poster

2017

NTCIR13 MedWeb Task: Multi-label Classification of Tweets using an Ensemble of Neural Networks
Hayate Iso, Camille Ruiz, Taichi Murayama, Katsuya Taguchi, Ryo Takeuchi, Hideya Yamamoto, Shoko Wakamiya and Eiji Aramaki
NII Testbeds and Community for Information Access Research (NTCIR), 2017
Best System Award [link]

Paper Slide Poster
BibTeX


@inproceedings{iso2017medweb,
  title = {NTCIR13 MedWeb Task: Multi-label Classification of Tweets using an Ensemble of Neural Networks},
  author = {Hayate Iso and Camille Ruiz and Taichi Murayama and Katsuya Taguchi and Ryo Takeuchi and Hideya Yamamoto and Shoko Wakamiya and Eiji Aramaki},
  booktitle={Proceedings of the 13th NTCIR Conference on Evaluation of Information Access Technologies},
  year={2017}
}

Density Estimation for Geolocation via Convolutional Mixture Density Network
Hayate Iso, Shoko Wakamiya and Eiji Aramaki
preprint, 2017

arXiv
BibTeX


@inproceedings{iso2017arxiv,
  title = {Density Estimation for Geolocation via Convolutional Mixture Density Network},
  author = {Hayate Iso and Shoko Wakamiya and Eiji Aramaki},
  journal   = {CoRR},
  volume    = {abs/1705.02750},
  year      = {2017},
  url       = {https://arxiv.org/abs/1705.02750},
  year = {2017},
}

2016

Forecasting Word Model: Twitter-based Influenza Surveillance and Prediction
Hayate Iso, Shoko Wakamiya, Eiji Aramaki
Computational Linguistics, (COLING), 2016
Acceptance rate: 337/1,039 = 32.4 %

Paper Slide
BibTeX


@inproceedings{iso2016forecasting,
  title = {Forecasting word model: Twitter-based influenza surveillance and prediction},
  author = {Hayate Iso and Shoko Wakamiya and Eiji Aramaki},
  booktitle = {Proceedings of International Conference on Computational Linguistics (COLING)},
  address = {Osaka, Japan},
  month = {December},
  pages = {76--86},
  url = {https://www.aclweb.org/anthology/C16-1008/},
  year = {2016},
}

Contact

444 Castro Street Suite 720 Mountain View, CA 94041
Megagon Labs