吉永 直樹 / Naoki Yoshinaga, PhD
For prospective students (updated on Nov. 4th, 2020): Due to enormous requests for research students and interns, I
ignore e-mails from free webmails (e.g., Gmail, Yahoo!, Outlook etc.) to filter out applicants who do not read the prospective supervisor's home page, while I
rarely reply to general-purpose applications that may be distributed to many laboratories as is to filter out applicants who do not seriously consider a matching between their own interests and research topics in the laboratory. We do not accept any students who
- wish to stay less than a half year as research intern without any concrete research plan. It's just too short to set up a task by themselves in academia. If you just want to enjoy fancy technology (e.g., deep learning) using existing NLP tasks as benchmark, you may want to consider applying to other laboratories or industry interns. Research requires time.
- wish to enter PhD program without having research experiences in NLP/CL fields, since such students were likely to take more than five years to complete PhD; I usually advise them to consider our research-oriented 2-year master program before entering to PhD program.
We yet accept a wide range of students who want to take master's or PhD program;
If you want to join our laboratories as master or PhD students, we strongly recommend you to visit us (Ee-503) talk via Zoom after having an appointment via e-mail. For those who want
to enter PhD program, send a presentation slide in English on a paper you have published in NLP/CL-field. For those who want
to enter master program, clarify whether you plan to proceed to PhD program.
Curriculum Vitae
2016 - present: | Associate Professor at Institute of Industrial Science, the University of Tokyo |
2014 - 2016: | Senior Researcher at National Institute of Information and Communications Technology (NICT) |
2012 - 2016: | Project Associate Professor at Institute of Industrial Science, the University of Tokyo |
2008 - 2012: | Project Assistant Professor at Institute of Industrial Science, the University of Tokyo |
2002 - 2008: | Research Fellow of the Japan Society for the Promotion of Science (JSPS) (DC1, PD) (11.6%, 10.0% accepted, respectively) |
2002 - 2005: | Ph.D. in Department of Computer Science, Graduate School of Information Science and Technology, the University of Tokyo |
2000 - 2002: | M.Sc. in Department of Information Science, Graduate School of Science, the University of Tokyo |
1996 - 2000: | B.Sc. in Department of Information Science, Faculty of Science, the University of Tokyo |
Full CV
Research Interests
We're studying various aspects on natural language processing (NLP) and computational linguistics (CL), especially
- Efficient algorithms and data structures for NLP [EMNLP-09, ACL-10, COLING-10, COLING-14]
- NLP in the wild; robust NLP with non-textual factors [IJCNLP-13, ACL-13, ACL-17 SRW, NAACL-19, F. EMNLP-20]
- NLP for breaking language barriers [CoNLL-15, ACL-17, NAACL-19, CoNLL-19, F. EMNLP-20]
- Mining from time-series social-media text [EMNLP-12, IJCAI-16, IJCAI-19]
- Information visualization for NLP [PacificVis-11, IUI-16, PacificVis-18]
I like to design novel and important NLP tasks [EMNLP-12, ACL-13, COLING-14, IJCAI-16, IJCAI-19], rather than solving classic tasks on worn-out datasets.
Research Grants
- NII CRIS Collaborative Research operated by NII CRIS and LINE Corporation (2020-2021): 13,000,000 JPY (Principal Investigator)
- NII CRIS Contract Research 2019 (2019-2020): 2,499,000 JPY (Principal Investigator)
- Grant-in-Aid for Scientific Research (B) (16H02905; 2016-2019): 17,810,000 JPY (as a collaborative researcher; Principal Investigator: Masashi Toyoda)
- U-Tokyo Excellent Young Researchers Start-up (2016-2018): 6,000,000 JPY (Principal Investigator)
- Grant-in-Aid for Young Scientists (B) (16K16109; 2016-2018): 3,900,000 JPY (Principal Investigator)
- IIS Sentei Kenkyuu (2016-2017): 2,000,000 JPY (Principal Investigator)
- Grant-in-Aid for JSPS Fellows (2005-2008) (Principal Investigator)
- Grant-in-Aid for JSPS Fellows (2002-2005) (Principal Investigator)
Softwares for NLP/CL
- opal: A scalable kernel-based online learner (polynomial kernel is supported).
- pecco: An efficient classifier for a model pre-trained with conjunctive features (or polynomial kernel).
- cedar: An efficient, updatable trie implementation based on double array.
- yakmo: A robust, efficient alternative k-means clustering.
- J.DepP: Very fast dependency parsers with the state-of-the-art accuracy for Japanese.
- RenTAL (no longer maintained): A grammar compiler from Lexicalized TAG to HPSG-style grammar.
Note that some of the above softwares are no longer
experimental codes; they are substantially elaborated from
the original codes to have a better performance (2-5x speed-up, or +1%
in accuracy etc.; track History section of each software). Those who want to reproduce the experimental results
may want to use the oldest release of the softwares.
Selected Publications (with links to presentations, codes, and datasets)
- Robust Backed-off Estimation of Out-of-Vocabulary Embeddings --- code
Findings of EMNLP-20 (long, acceptance rate 37%). Joint work with N. Fukuda and M. Kitsuregawa.
TL;DR: Inspired by two processes of creating words, we propose a simple word-based method to estimate OOV embeddings.
- Vocabulary Adaptation for Domain Adaptation in Neural Machine Translation --- code
Findings of EMNLP-20 (long, acceptance rate 37%). Joint work with S. Sato, J. Sakuma, M. Toyoda, M. Kitsuregawa.
TL;DR: We transplant target-domain vocabularies to source-domain NMT model for effecfive fine tuning.
- uBLEU: Uncertainty-Aware Automatic Evaluation Method for Open-Domain Dialogue Systems --- slide / code
ACL-20 SRW (acceptance rate 36%). Joint work with Y. Tsuta and M. Toyoda.
TL;DR: Our υBLEU exhibits higher correlation with human judgement than RUBER (AAAI-18).
- Multilingual model using cross-task embedding projection --- slide / code
CoNLL-19 (oral, acceptance rate: 22%). Joint work with J. Sakuma
TL;DR: Our locally-linear mapping optimizes multilingual models based on cross-lingual word embeddings to any tasks.
- On the Relation between Position Information and Sentence Length in Neural Machine Translation --- poster / code
CoNLL-19 (acceptance rate: 22%). Joint work with M. Neishi
TL;DR: We revealed that Transformer is bad at handling inputs of unseen lengths, and fixed it using RNN as relative position encoder.
- Early Discovery of Emerging Entities in Microblogs --- slide / poster / data
IJCAI-19 (acceptance rate: 13.7%). Joint work with S. Akasaki and M. Toyoda
TL;DR: We tackle a novel task of detecting emerging entities from Twitter timelines using timely distant supervsion.
- Learning to Describe Unknown Phrases with Local and Global Contexts --- slide / code & data
NAACL-19 (long, oral; acceptance rate 26%). Joint work with S. Ishiwatari, H. Hayashi, G. Neubig, S. Sato, M. Toyoda, and M. Kitsuregawa
TL;DR: Our LOG-CaD can explain unknown phrases; evaluated on new Wikipedia datasets.
- Modeling Personal Biases in Language Use by Inducing Personalized Word Embeddings --- slide
NAACL-19 (short, oral; acceptance rate 21%). Joint work with D. Oba, S. Sato, S, Akasaki, and M. Toyoda. (journal ver.)
TL;DR: We enable to model personalized usage of words, accompanied with analysis and applications.
- Information Integrated Visualization System for Heavy Rainfall Risk Analysis
PacificViz-18 (poster). Joint work with M. Itoh, T. Sagara, U. Suzuki, K. Umemoto, M. Toyoda, K. Zettsu, and Y. Kidawara
TL;DR: We integrate and visualize social sensor data to analyze heavy reaifall risks.
- A Bag of Useful Tricks for Practical Neural Machine Translation: Embedding Layer Initialization and Large Batch Size --- poster / code
WAT-17 (oral). Joint work with M. Neishi, J. Sakuma, S. Tohda, S. Ishiwatari and Masashi Toyoda
TL;DR: We confirmed the effectiveness of CBoW-based embedding layer initialization and Large Batch Size in NMT training.
- Modeling Situations in Neural Chat Bots --- poster
ACL-17 SRW (acceptance rate 36%). Joint work with S. Sato, M. Toyoda and M. Kitsuregawa.
TL;DR: We enable to model and incorporate user profiles and time in open-domain dialogue systems.
- Chunk-based Decoder for Neural Machine Translation --- poster
ACL-17 (long; acceptance rate 26%). Joint work with S. Ishiwatari, J. Yao, S. Liu, M. Li, M. Zhou, M. Kitsuregawa and W. Jia
Inspired by phrase-based SMT, we propose chunk-based decoding in NMT.
- Ordering Concepts Based on Common Attribute Intensity --- slide / poster / code & data
IJCAI-16 (acceptance rate <25%). Joint work with T. Iwanari, N. Kaji, T. Nishina, M. Toyoda and M. Kitsuregawa; a follow-up paper at COLING-16 (Demo) --- poster / software & data
TL;DR: We induce your sense of values from your writings; novel task.
- Spatio-temporal Event Visualization from a Geo-parsed Microblog Stream --- poster / demo by Prof. Itoh
IUI-16 (poster). Joint work with M. Itoh and M. Toyoda.
TL;DR: We visualize spatio-temporal tweets on word-clouds in the sky.
- Accurate Cross-lingual Projection between Count-based Word Vectors by Exploiting Translatable Context Pairs --- poster
CoNLL-15 (short, acceptance rate <30%). Joint work with S. Ishiwatari, N. Kaji, M. Toyoda and M. Kitsuregawa
TL;DR: We incorporate surface-based dimesional correspondences into count-based word vectors.
- A Self-adaptive Classifier for Efficient Text-stream Processing --- poster / software
COLING-14 (acceptance rate 32%). Joint work with M. Kitsuregawa
TL;DR: We propose a method of accelerating NLP classifiers when the processed text becomes redundant.
- Modeling User Leniency and Product Popularity for Sentiment Classification --- poster
IJCNLP-13. Joint work with W. Gao, N. Kaji, and M. Kitsuregawa. (journal ver.)
TL;DR: We enable to model annotation and selection biases in sentiment analysis for unseen users and targets.
- Predicting and Eliciting Addressee's Emotion in Online Dialogue --- poster
ACL-13 (long; acceptance rate 26%). Joint work with T. Hasegawa, N. Kaji, and M. Toyoda
TL;DR: We enable to model emotions in corpus-based open-dialogue systems.
- Identifying Constant and Unique Relations by using Time-Series Text --- slide
EMNLP-12 (oral; acceptance rate 17%). Joint work with Y. Takaku, N. Kaji, and M. Toyoda.
Use of massive Web in knowledge acquisition introduces enormous contradictions; solution provided.
- Analysis and Visualization of Temporal Changes in Bloggers' Activities and Interests --- demo / demo by Prof. Itoh
PacificVis-12 (acceptance rate 34%). Joint work with M. Itoh, M. Toyoda, and M. Kitsuregawa
TL;DR: We visualize dependencies in weblogs to understand the state of the world.
- Kernel Slicing: Scalable Online Training with Conjunctive Features --- slide / software & data
COLING-10 (oral; acceptance rate 19%). Joint work with M. Kitsuregawa
The kernel slicing generalizes kernel splitting (ACL 2008) to pack compuations in online learning with polynomial kernel.
- Efficient Staggered Decoding for Sequence Labeling --- software
ACL-10 (long; acceptance rate 25%). Joint work with N. Kaji, Y. Fujiwara, and M. Kitsuregawa
TL;DR: We made structured prediction scalable to the number of classes.
- Polynomial to Linear: Efficient Classification with Conjunctive Features --- poster / software & data
EMNLP-09 (acceptance rate 34%). Joint work with M. Kitsuregawa. (journal ver.)
TL;DR: We speed up testing of non-linear classifier with polynomial kernel.
- Boosting Precision and Recall of Hyponymy Relation Acquisition from Hierarchical Layouts in Wikipedia --- software
LREC-08. Joint work with A. Sumida and K. Torisawa
TL;DR: We developed a method of quickly obtaining a large-scale hyponymy relations from Wikipedia.
- Open-Domain Attribute-Value Acquisition from Semi-Structured Texts --- demo
ISWC-07 workshop, OntoLex. Joint work with K. Torisawa; a follow-up paper on writing-support environment
TL;DR: We developed a method of extracting attributes and their values for a given object using unsupervised wrapper induction.
- Improving the Accuracy of Subcategorizations Acquired from Corpora
ACL-04 SRW (acceptance rate 28%)
- A Debug Tool for Practical Grammar Development
ACL-03 (poster; acceptance rate 43%). Joint work with A. Yakushiji, K. Tateisi, Y. Miyao, and J. Tsujii
- Comparison between CFG Filtering Techniques for LTAG and HPSG
ACL-03 (poster; acceptance rate 43%). Joint work with Y. Miyao, K. Torisawa, and J. Tsujii (journal ver.)
- Grammar Conversion from LTAG to HPSG --- software
ESSLLI 2011 Student Session (oral; acceptance rate 26%). Joint work with Y. Miyao. (journal ver.)
More publications
Awards
- Best Interactive Award (1st place): the eighth Forum on Data Engineering and Information Management (DEIM) (2019)
- JSAI 30th Anniversary Best Paper Award (2016)
- Best Poster Award: the 17th International Conference on Intelligent Text Processing and Computational Linguistics (CICLing) (2016)
- Best Interactive Award: the eighth Forum on Data Engineering and Information Management (DEIM) (2016)
- Best Paper Award: the WebDB Forum (2015)
- Business Award (Yahoo! Japan Award): the WebDB Forum (2014)
- JSAI SIG Research Award 2013: Japanese Society for Artificial Intelligence, Special Interest Group on Fundamental Problems of Artificial Intelligence (SIG-FPAI) (2013)
- Best Student Paper Award: IEICE Transactions on Information and Systems (2013).
- Best Paper Award: the third Forum on Data Engineering and Information Management (DEIM) (2011).
- Best Paper Award: the 72nd National Convention of IPSJ (2010).
- Best Paper Award: Journal of Natural Language Processing (2009).
- Business Award (1st place): the Symposium on DataBases and Web Information Systems (DBWeb) (2007)
Awards (as a supervisor):
- Young Researcher Award: the 26th Annual Meeting of the Association for Natural Language Processing (NLP) (2020).
- Young Researcher Award: the 25th Annual Meeting of the Association for Natural Language Processing (NLP) (2019).
- Student Presentation Award: the seventh Forum on Data Engineering and Information Management (DEIM) (2019), for two papers.
- Student Presentation Award: the seventh Forum on Data Engineering and Information Management (DEIM) (2015).
- Student Incentive Award: the WebDB Forum (2014)
- Student Presentation Award: the sixth Forum on Data Engineering and Information Management (DEIM) (2014).
- Student Incentive Award: the WebDB Forum (2013)
- Young Researcher Award: the 19th Annual Meeting of the Association for Natural Language Processing (NLP) (2013).
- Student Presentation Award: the fifth Forum on Data Engineering and Information Management (DEIM) (2013).
- Best Student Paper Award: IEICE technical report on Natural Language Understanding and Models of Communication (2012).
- Student Incentive Award: the fourth Forum on Data Engineering and Information Management (DEIM) (2012).
last-modified: Mar 07 17:22:17 2021, written by
XHTML 1.1