Assistant Professor
McGill NLP Group
Mila -- Quebec Artificial Intelligence Institute
School of Computer Science, McGill University
Department of Linguistics, McGill University
Researcher at ServiceNow Research
I am an Assistant Professor in the School of Computer Science and Linguistics at McGill University, a Facebook CIFAR AI Chair, a core faculty member of Mila Quebec AI Institute, and a research scientist at ServiceNow Research.
I co-lead the McGill NLP Group.
Previously, I was a postdoctoral researcher at Stanford University and a Google PhD fellow at the University of Edinburgh.
My research interests are in representation learning for language with a specific focus on reasoning, grounding, conversational modeling, and safety.
Broadly, my research has four themes:
Theme I: How can we represent language so that machines can effectively process and reason upon it? This quest also involves a scientific pursuit of language: if one representation is better than the other, can we gain insights into why the underlying representation is better and what it can teach us about linguistic structure?
Theme II: Can we interact with machines the way we interact with people to get tasks accomplished, such as having a conversation to seek answers about a topic of interest or asking a robot to perform a task? I have been working on methods that improve trustworthiness of these models using retrieval-augmentation
Theme III: Even if we succeed at Themes I and II, it is important to ensure that the resulting advances are safe. Additionally, it is important that these advances are available in several languages, not just in English.
Theme IV: Language is often situated with rich modalities such as vision and speech, and humans thrive in such environments. How can we build language models that integrate these modalities?
I am actively looking for highly motivated PhD and MSc (thesis) students. For prospective students, here is a list of topics I work on.
Theme I: How can we represent language so that machines can effectively process and reason upon it? This quest also involves a scientific pursuit of language: if one representation is better than the other, can we gain insights into why the underlying representation is better and what it can teach us about linguistic structure?
Theme II: Can we interact with machines the way we interact with people to get tasks accomplished, such as having a conversation to seek answers about a topic of interest or asking a robot to perform a task? I have been working on methods that improve trustworthiness of these models using retrieval-augmentation
Theme III: Even if we succeed at Themes I and II, it is important to ensure that the resulting advances are safe. Additionally, it is important that these advances are available in several languages, not just in English.
Theme IV: Language is often situated with rich modalities such as vision and speech, and humans thrive in such environments. How can we build language models that integrate these modalities?
I am actively looking for highly motivated PhD and MSc (thesis) students. For prospective students, here is a list of topics I work on.
Quick links: CV, Google Scholar, Twitter, Github.
Postdocs
Marius Mosbach (Postdoc Mar 2024)Karolina Stańczak (Postdoc Mar 2024)
PhD Students
Vaibhav Adlakha (Sep 2020)
Zichao Li (Sep 2020; co-advised with Jackie Cheung)
Nicholas Meade (Sep 2020)
Benno Krojer (Sep 2021)
Xing Han Lu (Sep 2021)
Parishad BehnamGhader (Jan 2022)
Gaurav Kamath (May 2022; Linguistics Eval 1 and 2)
Arkil Patel (Sep 2022; co-advised with Dzmitry Bahdanau)
Mehar Bhatia (Sep 2024; co-advised with Vered Shwartz)
Aditi Khandelwal (Sep 2024; co-advised with Golnoosh Farnadi)
Masters Students
Research Assistants/Interns
Amirhossein Kazemnejad (Dec 2023; Graduated with MSc from the group)
Alumni
Andreas Madsen (PhD Sep 2020--Nov 2024 -> Founding Research Scientist @ Guide Labs; co-advised with Sarath Chandar)
Eva Portelance (Postdoc Jan 2023 -> Assistant Professor at HEC Montreal; co-advised with Tim O'Donnell)
Aristides Milios (MSc Sep 2021 -> PhD at Polytechnique Montreal; co-advised with Dzmitry Bahdanau)
Zdeněk Kasner (Intern Feb 2023 -> Back to Charles University)
Jasper Jian (Undergrad; 2022/23 -> PhD at Stanford Linguistics)
Sivan Milton (Undergrad; 2021/2022; -> PhD at Edinburgh NLP)
Elinor Poole-Dayan (Undergrad; Fall 2022; Winter 2021/23 -> MS at MIT Media Lab)
Nathan Schucher (MSc Sep 2020 -> Google DeepMind)
Laurestine Bradford (Jan 2022; Linguistics Eval 1; co-advised with Tim O'Donnell -> back to McGill)
Xuanda Chen (Sep 2022; Linguistics Eval 2; co-advised with Tim O'Donnell -> back to McGill)
Edoardo Ponti (Postdoc; Oct 2020 -> Assistant Professor at the School of Informatics, University of Edinburgh)
Emanuele Bugliarello (Intern Jan 2022 -> Intern at Google DeepMind)
Nouha Dziri (Visting Student May 2021 -> Postdoc at AI2)
Devang Kulshreshtha (MSc Sep 2020 -> Applied Scientist, Amazon AWS)
Neshma Metri (Cognitive Science Honors Fall 2021)
Michelle Yang (Undergrad Fall 2021 -> back to McGill; co-advised with Morgan Sonderegger)
Emily Goodwin (MA -> PhD at Stanford Linguistics; Jan 2020; co-advised with Tim O'Donnell)
Lucas Torroba Hennigen (Intern -> PhD at MIT Computer Science)
Victor Prokhorov (Intern, back to Cambridge University)
Service
Senior area chair: EMNLP 2024
Ethics area chair: COLM 2024
Tutorial chair: ACL 2023
Senior area chair: NAACL 2022
Area chair: EMNLP 2022
SRW faculty advisor: ACL 2022
Area chair: NAACL 2021
Area chair: EMNLP 2020
Area chair: ACL 2020
Area chair: ACL 2019
Education
Post-doctoral researcher at Stanford University with Chris Manning.
PhD from University of Edinburgh with Mirella Lapata and Mark Steedman.
Past Industry Experience
Internship at Google NLU
Engineer at Sketch Engine