Rate My Professor Shareghi Nojehdeh

Shareghi Nojehdeh obtained his PhD in 2017 from the Department of Information Technology at Monash University (Clayton campus), with a thesis entitled 'Scalable Non-Markovian Sequential Modelling for Natural Language Processing.' This work demonstrated that finite-order Markov models inadequately capture long-range dependencies in human language and introduced scalable alternatives for improved sequential modelling in NLP tasks. Since joining Monash in 2021, he has held positions up to Senior Lecturer. His research applies AI and machine learning to diverse domains, serving as Chief Investigator on initiatives like 'Grid Guru: Leveraging AI-Driven Grid Optimisation' under the Monash Energy Institute and 'Tracking Media and Parliamentary Narratives around Cycles of Disaster' in the Department of Data Science and AI. He co-supervises student projects such as 'Learning from Massive Amounts of EEG Data.' Key publications from his Monash period include 'Self-alignment Pre-training for Biomedical Entity Representations' (2021, 530 citations), 'COMETA: A Corpus for Medical Entity Linking in the Social Media' (2020, 127 citations), 'A Closer Look at Few-Shot Crosslingual Transfer: The Choice of Shots Matters' (2021, 70 citations), 'TaCL: Improving BERT Pre-training with Token-aware Contrastive Learning' (2022, 72 citations), and 'Mixture-of-Partitions: Infusing Large Biomedical Knowledge Graphs into BERT' (2021, 50 citations). These works, often in collaboration with researchers like Fangyu Liu, Zaiqiao Meng, Marco Basaldella, and Nigel Collier, have amassed over 2,200 citations on Google Scholar, influencing advancements in biomedical NLP, entity linking, and language model pre-training. His interdisciplinary efforts extend to energy optimisation and social narrative analysis.