Hady Elsahar | Naver Labs Europe
Change your cover photo
Change your cover photo
This user account status is Approved

This user has not added any information to their profile yet.

Natural Language Processing

I am a researcher Scientist in Naver Labs Europe since Jan. 2019.
I am particularly interested in Natural Language Generation and specifically under non-traditional conditions, including:

- Unsupervised Abstractive Summarization
- Zero-Shot NLG
- Data2text for low-sourced languages
- Controlled Natural Language Generation (available internship position)
- Constrained decoding


When I am not working on Generation I do research activities on domain-shift detection (check our latest EMNLP2019 paper) and I work on the SMOOTH EU project for GDPR compliance detection.



For the full list of my publications check my Google Scholar or DBLP
For more details check my personal website: hadyelsahar.io

I hold a PhD. Degree from L'Université de Lyon since 2019. My thesis was on enhancing evolving interactions between Natural Language and Structured Knowledge bases to allow applications such as Data2Text Generation and Relation Extraction & Discovery
Prior to that, I got a Masters Degree from Nile University in Cairo where I worked on Low-resource sentiment analysis. 

Previous to Naver Labs Europe I did several research internships in research teams in Bloomberg London, IBM R&D Labs in German and Microsoft Research Labs in Cairo. 


I believe that research cannot be disassociated from the society we live in, Therefore, every researcher should keep themselves aware of social biases such as accessibility and language

  • Research Internship Position: Reinforcement Learning for Controlled Text Generation . Contact me for more details or Apply Here!
  • EMNLP2019 Paper On prediction of performance drop under Domain-Shift.  [paper] [blogpost]
  • EMNLP 2019 W/S paper accepted on Unsupervised Abstractive Summarization in New Frontiers in Summarization workshop.