cv
General Information
| Full Name | Nadir Durrani |
| Languages | English, Urdu, Hindi, Punjabi, German (elementary) |
Education
-
2008-2012
PhD (Dr. rer. nat.) with Magna Cum Laude
University of Stuttgart, Stuttgart, Germany
- A Joint Translation Model with Integrated Reordering
- Prof. Hinrich Schütze, Prof. Alexander Fraser, Dr. Helmut Schmid
- GSCL Best Thesis Award (2014)
-
2005-2007
MS (CS) with Silver Medal
National University of Computer and Emerging Sciences, Lahore, Pakistan
- Typology of word and automatic word segmentation in Urdu text corpus
- Prof. Sarmad Hussain
Experience
-
2021 - date
Senior Scientist
Qatar Computing Research Institute, Doha, Qatar
- LLMeBench: Benchmark your LLMs
- NxPlain: Explain my prediction
- NeuroX: A framework to interpret deep NLP models
- NatiQ: Arabic Speech Synthesis System
-
2015 - 2020
Scientist
Qatar Computing Research Institute, Doha, Qatar
- Shaheen: Dialectal Arabic-to-English System
- Shaheen: MSA Arabic-to-English-to-Arabic System
- QATS: Arabic-English Speech Translation System
- NeuroX: A framework to interpret deep NLP models
-
2012-2014
Postdoctoral Research Associate
ILCC, University of Edinburgh, Scotland, UK.
- Translation and Reordering Models for Phrase-based MT
- Unsupervised transliteration models in SMT
-
2012-2012
Research Intern
IBM Watson Center, Yorktown Heights, NY, USA
- Egyptian-to-English Machine Translation
-
2005-2007
Research Officer
Center for Research in Urdu Language Processing, Lahore, Pakistan
- PAN Localization Project
-
2005-2005
Project Coordinator
Science and Technology and Environment Agency, Vientiane, Laos
- PAN Localization Project
Projects
-
2023-
LLMeBench
- A framework to benchmark LLMs
-
2021-2023
NxPlain
- A framework to explain predictions in NLP models
-
2018-
NeuroX
- A framework to interpret deep NLP models
-
2018-2021
NatiQ
- NatiQ is an Arabic Text-to-Speech Synthesizer based on encoder-dercoder architecture
-
2015-2021
Shaheen
- Shaheen is an Arabic-English Neural Machine Translation system with support of various Arabic dialects
-
2015-2016
Farasa
- Farasa is the state-of-the-art full-stack package to deal with Arabic Language Processing
-
2012-2015
Moses
- Moses a statistical machine translation system
Honors and Awards
-
2023
- King Salman Global Academy Prize, Riyadh, Saudi Arabia
-
2017
- Best innovation award for Arabic-English Speech Translation at ARC, Doha
-
2014
- GSCL Best Computational Linguistic Dissertation Award, Kovens, Germany
-
2007
- Scolarship for doctoral studies in Germany, Islamabad, Pakistan
Research Interests
-
Natural Language Processing
- Explainable AI: Analyzing deep NLP models (with a focus on diagnostic classifiers, linguistic correlation analysis, unsupervised methods, clustering methods, ablation analysis, similarity analysis, causation and behavioral studies)
- Machine Translation: Neural and Statistical MT (with focus on reordering, domain adaptation, transliteration, dialectal translation, pivoting, closely related and morphologically rich languages, simultaneous translation, eye-tracking for MT evaluation)
-
Localization
- Localization of open source operating systems (desktop environment KDE/GNOME), word processing applications (Open Office), chatting tools (Psi), web browsers (Sea Monkey, Mozilla) and web development tools (NVU), localization of domain names, font development