Scholar
Nadir Durrani
Google Scholar ID: K6uisFAAAAAJ
Senior Scientist, QCRI, HBKU
Citations & Impact
Citations
4,091
H-index
29
i10-index
47
Publications
20
Co-authors
131
list available
Publications
Resume (English only)
Academic Achievements
- Two papers accepted to EMNLP 2025; one paper 'From Words to Waves: Analyzing Concept Formation in Speech and Text-Based Foundation Models' accepted for presentation at INTERSPEECH 2025. Published research on discovering salient neurons in deep NLP models in the Journal of Machine Learning Research.
Research Experience
- Works at QCRI on various projects including NeuroX, Shaheen, NatiQ, and Farasa. Previously, worked as a Research Associate at the Institute of Language, Cognition and Computation, University of Edinburgh, focusing on problems in SMT such as Unsupervised Transliteration and Markov-based translation models.
Background
- Senior Scientist at the Arabic Language Technologies (ALT) working on Interpretability (NeuroX), Machine Translation (Shaheen), Speech Synthesis (NatiQ), and language processing tools for Arabic (Farasa). Previously, a Research Associate under Philipp Koehn at the Institute of Language, Cognition and Computation, University of Edinburgh.
