Courses
-
In 2023 fall academic semester I helped lecture the undergraduate course Fundamentals of Machine Learning together with Prof. Dr. Sumit Chopra. For the practical component of the course, I designed and implemented practical learning material teaching fundamental methodologies for implementing machine learning models with Pytorch, tuning hyper-parameters, and experimentation across various model designs, choices of data sets, and various regularization or optimization methods. Interested parties can contact me to access the learning material.
-
During the 2022 spring-fall academic semesters I taught the undergraduate course Introduction to Computer Science at the Computer Science program at the New York University, which provides an introduction to object-oriented programming with Java. The material from my online course in fall 2022 is accessible online through the public website for aiding the self-learning experience of all Java enthusiasts.
-
I have been a visiting lecturer at the Natural Language Processing program at the African Master’s in Machine Intelligence (AMMI) where I co-instructed a seminar together with Prof. Dr. Kyunghyun Cho. The slides from my lectures on machine translation and multilinguality can be found online at the following links.
-
In 2020, I prepared and taught a new course introducing data science to bachelor's students in computational linguistics at the University of Zürich. The course provides an introduction to resources for text and speech processing, with a particular focus on corpus linguistics, manual and automatic annotation methods, statistical sampling methods, and data annotation and processing libraries using Python for curating, cleaning and annotating corpus resources.
Below you can find all course material from the lectures:
Supervision
During the time I was at the University of Zürich and New York University, I mentored many students that either perform individual (thesis) research under my supervision, or contribute to the ongoing projects in my research group. I am proud to present some of my current and previous students that successfully completed (or close to completion) research projects under my supervision.
Raghav Mantri, M.Sc. New York University
2024 - Ongoing
Permutation equivariant models and their integration into multilingual language models for mitigating position bias
Jonne Saleva, Ph.D. Brandeis University (Visiting post-graduate student)
2024
Statistical evaluation methods for reliable comparison of generative language models across languages and domains
Saksham Bassi, M.Sc. New York University
2023 - 2024
Development of parametric methods for measuring generalization capability of large language models in previously unseen languages
Tia Chen, B.Sc. Tufts University (Pathways to AI Scholar)
2022 - 2023
Evaluating the generalization performance of automatic evaluation metrics to rephrasing across languages
Saun Chen, B.Sc. New York University
2024
Investigating geometric properties of underlying mechanisms enabling cross-lingual transfer in large language models
Jafar Isbarov, M.Sc. George Washington University (Thesis)
2024 - Ongoing
Statistical prompt compression methods for increasing efficiency of generative language models
Vivien Angliker, M.Sc. University of Zürich (Thesis)
2020 - 2021
Multi-lingual extractive text summarization models for structural adaptation across languages
Francesco Tinner, B.Sc. University of Zürich (Thesis)
2021 - 2022
Investigating zero-shot cross-lingual generalization in the multilingual topic modeling task using large language models
Sushma Mareddy, M.Sc. New York University
2024 - Ongoing
Developing novel methods for automatic evaluation of generated language
Grace Wang, B.Sc. New York University
2022 - 2023
Investigation of evaluation metrics applicable for language generation across languages
Cecilia Zhang, B.Sc. New York University
2024
Non-linear manifold learning for analysis of language-specific representation spaces in multilingual large language models
Zijian Jin, M.Sc. New York University
2022 - 2023
Design of multi-modal language models for improving inference tasks in graphemic languages