280px-laboratoire_lorrain_de_recherche_en_informatique_et_ses_applications_logo-svg        cnrs_fr_quadri-svg

I’m currently Ph.D. student (2016-2020) at LORIA (a group of research CNRS, INRIA, University of Lorraine), working in collaboration with Christophe Cerisara and Claire Gardent (CNRS) (*). My research interests are neural models with focus on summarization, generation and sentiment analysis tasks. During summer and winter 2017, I spent 6 months at Google Research Europe in Zurich, Switzerland, working on text-to-text generation and abstractive summarization. In spring 2018, I taught Machine Learning course for Master Sciences de la Cognition et Applications (Erasmus Mundus) of University of Lorraine and mentored for Summer School in NLP in 2019. I’m also reviewer for some NLP conferences (below). Throughout the Ph.D. and internships, I found that I really love doing research, writing, engineering, solving problems and especially teaching !

Before that, I graduated Master Stochastic Modeling (Data Science track) at University Paris 1 Pantheon Sorbonne and University of Paris Diderot, France. I  also followed a Master of Quantitative Economics & Finance at Ecole Polytechnique & HEC Paris. I’m originally from Vietnam (and soon will become French) though I also feel extremely lucky to have lived in these beautiful cities Paris, Nancy and Zurich.

Contact: hoa.le __at__ loria __dot__ fr

My life before coming to France: CV

Also: Google Scholar, Github

(*) : If you want do a Ph.D in France or especially in the Grand Nord-East region of France, I would like to recommend you to Nancy if you want to enjoy a good time of doing the Ph.D !

(**) : I particularly believe in the (bright) future of Deep Learning applications on Mobile Devices. If you’re also interested, you can follow more here.


Academic Activities:

Participated in conference, events:

  • La journée “IA, Langage et Citoyens” – OLKi project. University of Lorraine
  • PhD Student Day APIL 2018. University of Lorraine [Poster]
  • EMNLP 2018. Brussels
  • International Workshop of Machine Learning & Artificial Intelligence 2018. Telecom ParisTech
  • Deep Learning & Reinforcement Learning Summer Schools 2018 (DLRL) at Vector Institute (travel & registration fee grant). Toronto
  • Seminar Artificial Intelligence of French-American Doctoral Exchange (FADEx 2018)  (among 10 AI phd students represent France; travel, accommodation, food grant) [Poster]Nice, Grenoble, Paris
  • AAAI 2018. New Orleans, USA
  • Google Natural Language Processing Summit 2017. Zurich
  • Ecole Polytechnique Data Science Summer School 2017 [Recap VAE – GAN] [Poster]. Paris
  • ICLR 2017. Toulon, France
  • NIPS 2016. Barcelone


  • INLG 2019
  • ACL 2019
  • SIGDIAL 2017


Teaching experiences:

Summer 2019

Spring 2018

  • Teaching Memory and Machine Learning Course of Master Sciences de la Cognition et Applications (University of Lorraine, Master Erasmus Mundus) [Materials]
  • Co-supervision internship of Master student Ecole de Mines Nancy. Topic ¨Abstractive Summarization with Variational Seq2seq¨

Summer 2017

  • Co-supervision internship of Master SCA Erasmus Mundus. Topic ¨Dialogue Act and Sentiment Analysis on Twitter using Memory Networks¨.


  • “Quality of syntactic implication of RL-based sentence summarization”.  Hoa T. Le, Christophe Cerisara, Claire Gardent. Workshop on Engineering Dependable and Secure Machine Learning Systems of the AAAI-20 Conference on Artificial Intelligence (Oral presentation). New York, America. [Paper]
  • “RL extraction of syntax-based chunks for sentence compression”.  Hoa T. Le, Christophe Cerisara, Claire Gardent. International Conference on Artificial Neural Networks (ICANN-19) (Oral presentation). Munich, Germany. [Paper][Slides][HAL Archives]
  • “Multi-task dialog act and sentiment recognition on Mastodon”. Christophe Cerisara, Adedayo Oluokun, Somayeh Jafaritazehjani, Hoa T. Le. The International Conference on Computational Linguistics (COLING-18) (Acceptance rate: 37.4%). Santa Fe, New Mexico, USA.[Paper] [Code]
  • “Do Convolutional Networks need to be Deep for Text Classification ?” Hoa T. Le, Christophe Cerisara, Alexandre Denis. Workshop on Affective Content Analysis of the AAAI-18 Conference on Artificial Intelligence (Oral presentation) (Acceptance rate: 20.68%). New Orleans, US. [Paper] [Poster] [Slides] [Code]


  • Research sharing talk at VietAI. Aug 2018. Ho Chi Minh city, Vietnam
  • Organizing and maintaining Reading Group of Deep Learning & Natural Language Understanding in Synalp team. Feb-May 2018.
  • A review of NIPS 2016 – DeepLoria. Jan 2017.
  • Is Very Deep Convolutional Neural Network necessary for Text Classification? – Synalp. Nov 2016.


Fellowships & Achievements:

  • Phd Fellowship of Lorraine Université d’Excellence (LUE). 2016.
  • Scholarship (two consecutive years) of Centre France – Asie (CFA). 2014-2015.
  • Master Fellowship to follow Statistics at Toulouse School of Economics by Agence Universitaire de la Francophonie (AUF) (renounce). 2013.
  • Bachelor top 5 GPA of the Mathematics – Statistics department, University of Economics HCM city. Second rank bachelor’s thesis of francophone program (defense in French). 2013.
  • Third Prize of National Olympic Programming Competition for Students (Vietnam). 2011.
  • First prize (two times), Second prize (one time) of the Informatics competition, University of Economics HCM city. 2009-2011.
  • Top 30 Vietnam in final round national contest Microsoft Office World Champion. 2010.
  • Highest entrance GPA of the Mathematic – Statistics Department, University of Economics Ho Chi Minh city. 2010.