Computer Science Seminar Series
Presenter: Dr. David Chandross (Department of Computer Science, Toronto Metropolitan University and Adjunct Professor, Game Development and Interactive Media Program, Ontario Tech University)
Host: Dr. Bill Kapralos (Faculty of Business and Information Technology, Ontario Tech University)
Title: Tripping the Light Fantastic, Psychedelic Science Applied to Teaching
Abstract: In this presentation we will explore how recent research on psychedelic psychotherapy impacts our views on the neuroscience of learning.
Presenter Bio: David Chandross holds a Ph.D. in curriculum for medical education, a M.Sc. in cognitive neuroscience, a M.Ed. in medical education. He has just completed 4 years of working with the WHO and the UN on game based learning. His clients include Baycrest Health Science, Alter Spark, U of T, CDN Armed Forces, Insurance Institute of Canada and other similar organizations.
-
Past Seminars
Date: Wednesday, October 23, 2024
Time: 1:00 pm – 2:00 pm
Room: UA 4170 (Science Board Room)Presenter: Dr. Freda Shi (David R. Cheriton School of Computer Science, University of Waterloo, Waterloo, Canada.)
Host: Dr. Annie En-Shiun Lee (Faculty of Science, Ontario Tech University)
Title: Towards computational multilingualism with large language models
Abstract: In light of the remarkable accomplishments of recent large language models (LLMs), there arises a natural question on their multilingual capability, especially regarding low-resource and underrepresented languages. In this talk, I will discuss our results on cross-lingual reasoning with LLMs, where we investigate how pretrained models can be prompted to reason in a cross-lingual manner. Next, I will present our work that leverages multilingual LLMs for lexicon induction and, time permitting, cross-lingual dependency parsing. In line with existing literature, on the one hand, we show that the pretrained multilingual models benefit pre-defined cross-lingual NLP tasks across the board; on the other hand, the experimental results lead to a deeper understanding of what happens within these pretrained models.
Presenter Bio: Freda Shi is an Assistant Professor in Computer Science at the University of Waterloo, and a Faculty Member at the Vector Institute, where she also holds a Canada CIFAR AI Chair. She received her PhD in computer science from the Toyota Technological Institute at Chicago. Her research interests are in computational linguistics and natural language processing, as well as relevant machine learning aspects. Her work has been recognized by nominations for the Best Paper Award at ACL 2019, 2021 and 2024 and a Google PhD Fellowship.
Date: Wednesday, October 23, 2024
Time: 4:00 pm – 5:00 pm
Location: Online (Zoom) https://utoronto.zoom.us/j/81989721154Presenter: Dr. Vered Shwartz (Department of Computer Science, University of British Columbia, Vancouver, Canada, and CIFAR AI Chair, Vector Institute, Toronto, Canada)
Host: Dr. Annie En-Shiun Lee (Faculty of Science, Ontario Tech University)
Abstract: Despite their amazing success, large language models and vision and language models suffer from several limitations. This talk focuses on one of these limitations: the models’ narrow Western, North American, or even US-centric lens, as a result of training on web text and images primarily from US-based users. As a result, users from diverse cultures that are interacting with these tools may feel misunderstood and experience them as less useful. Worse still, when such models are used in applications that make decisions about people’s lives, lack of cultural awareness may lead to models perpetuating stereotypes and reinforcing societal inequalities. In this talk, I will present a line of work from our lab aimed at quantifying and mitigating this bias.
Bio: Vered Shwartz is an Assistant Professor of Computer Science at the University of British Columbia, and a CIFAR AI Chair at the Vector Institute. Her research interests include commonsense reasoning, computational semantics and pragmatics, multiword expressions, and multimodal models. Previously, Vered was a postdoctoral researcher at the Allen Institute for AI (AI2) and the University of Washington, and received her PhD in Computer Science from Bar-Ilan University.
Date: Wednesday, October 2, 2024Time: 1:00 pm – 2:00 pm
Room: UA 4170 (Science Board Room)
Presenter: Dr. Jessie Galasso (Department of Electrical and Computer Engineering at the Faculty of Engineering, McGill University, Canada.)
Host: Dr. Cristiano Politowski (Faculty of Science, Ontario Tech University)
Title: Navigating the challenges of creating large, reproducible software datasets
Abstract: The collection and management of software datasets are fundamental to empirical software engineering. Defining and adopting proper data collection practices is essential to ensure the quality and reproducibility of research findings. In this talk, I will explore the challenges and limitations researchers face when selecting repositories from popular code forges. Specifically, I will address the issues that may hinder the reuse and reproducibility of software datasets. Following this, I will present solutions to support the creation and sharing of large, reproducible datasets of software repositories. These solutions are centered around the concept of dataset fingerprinting and leverage the unique features of the Software Heritage open archive.
Presenter Bio: Jessie Galasso is an Assistant Professor in the Department of Electrical and Computer Engineering at the Faculty of Engineering, McGill University, Canada. She received a Ph.D. in Computer Science in 2018 from the University of Montpellier, France. Her primary research interests focus on issues related to the management of software datasets, including software repository mining, knowledge extraction and representation, as well as variability management and modeling. In recent years, she has served as a program committee member for SANER'24 and ICSR'24, and has co-chaired several tracks and workshops at MODELS and SPLC.
