Hassan Sajjad

Hassan Sajjad is an Associate Professor in the Faculty of Computer Science at Dalhousie University, Halifax, Canada, a position he has held since August 2022. He serves as the Director of the HyperMatrix Research Lab, which conducts research in artificial intelligence with a focus on natural language processing, deep learning, and safe and trustworthy AI. His work explores critical areas such as interpretability and explainability of neural networks, generalization and robustness, language generation, model editing, steering, compositionality, conformal prediction, and LLM agents. Prior to Dalhousie, Sajjad was Senior Research Scientist at the Qatar Computing Research Institute (QCRI), HBKU, from January 2021 to August 2022, where he managed projects on machine translation and neural network interpretation. He previously served as Research Scientist at QCRI from 2014 to 2020, leading machine translation development and commercialization, including collaborations with MIT and the H2020 SUMMA project, and as Post-doctoral Researcher at QCRI from 2013 to 2014 on dialectal Arabic machine translation. Sajjad earned his PhD in Computer Science (Magna Cum Laude) from the University of Stuttgart, Germany, in 2012, with a thesis on statistical models for unsupervised, semi-supervised, and supervised transliteration mining. He holds an MS (2007) and BSc (2005) from the National University of Computer and Emerging Sciences, Pakistan, and interned at Microsoft Research in 2011.

Sajjad's scholarly impact is evident in key publications such as "Discovering Salient Neurons in deep NLP models" (Journal of Machine Learning Research, 2023), "Neuron-level Interpretation of Deep NLP Models: A Survey" (Transactions of the Association for Computational Linguistics, 2022), "What is one Grain of Sand in the Desert? Analyzing Individual Neurons in Deep NLP Models" (AAAI, 2019), and "On the Effect of Dropping Layers of Pre-trained Transformer Models" (Computer Speech & Language, 2023). His technologies have been deployed to BBC and Deutsche Welle, earning media coverage from MIT News and Gulf Times. Awards include the Best Innovation Award at Qatar's Annual Research Conference 2018 for a speech translation system and Outstanding Reviewer at EMNLP 2020. He has received grants like the NSERC Discovery Grant (CAD 145,000, 2022-2027), Research Nova Scotia (CAD 148,553, 2023), and CFI JELF (CAD 148,553, 2022). At Dalhousie, he teaches Deep Learning for Natural Language Processing and Introduction to Experimental Robotics, and has delivered international crash courses and served as a panelist at conferences.