Learning & Education Advancement Fund+ Project Theme: Generative Artificial Intelligence in Teaching and Learning (LEAF+AI)
The LEAF+ AI is a new program funded by the Vice-Provost Innovations in Undergraduate Education (OVPIUE) that addresses the theme of generative AI in teaching and learning. Given the potential impacts, opportunities and challenges of this emergent technology for the ways our course instructors and students engage in learning, the OVPIUE is interested in projects to help inform the community’s use in pedagogical contexts.
The successfully funded recipients have formed a cross-disciplinary network and will meet periodically to share insights as new approaches using or responding to AI in teaching and learning are prototyped across the projects. This early exploration will serve as a conduit for feedback to OVPIUE on instructors’ experiences piloting new strategies and methods.
Read full details about the projects below:
AI for Image Research in Art and Architecture
Deployment of LLM-based Personal Coding Assistants that Balance Helpfulness and Directness
Designing Differentiated Instruction in the Foreign Language Classroom Using ChatGPT
Exploring Ethical and Creative Uses of Generative AI to Support Equity-Deserving Students
Instructor-AI Collaborative Content Generation Project
Learning to Teach and Teaching to Learn in the Generative AI Landscape
Strengthening Co-op Student’s Communication Skills for Job Search and Workplace Success
Understanding the Limits of AI-Based Image Generators with DALL·E 2 and Midjourney
AI for Image Research in Art and Architecture
Project Lead: Cathryn Copper, University of Toronto Libraries, Eberhard Zeidler Library
This project seeks to educate budding artists and architects on how to ethically use AI in image research, how research outcomes differ between AI and traditional research methods, and how to think critically about AI generated images. Instructional materials will be created and offered in first year undergraduate courses around these general topics:
- How to ethically use AI in image research – Introduce image generation software, discuss how algorithms work, and combat plagiarism
- How AI and traditional research outcomes – Demonstrate image research with AI tools and traditional methods, discuss differences in results, share how to use AI for research synthesis
Deployment of LLM-based Personal Coding Assistants that Balance Helpfulness and Directness
Project Lead: Tovi Grossman, Faculty of Arts & Science, Computer Science
Existing LLM-based tools like ChatGPT are designed to generate direct responses to programming questions and produce code. Such directness may give away solutions without making the learner think deeply and potentially compromise academic integrity. In this project, we developed an LLM-based coding assistant that generates immediate and personalized responses designed in a way to meet the criteria of both students and educators: being helpful, technically correct, while not directly revealing the code solution. We deployed the assistant in a large introductory programming class to (i) explore students’ usage of the assistant including what questions they ask and how they are framed, (ii) thematically analyze the assistant’s responses to questions on correctness, helpfulness, explainability, and directness, (iii) interview and survey students to understand their perspectives about the assistant and compare its usefulness to other resources including ChatGPT, and (iv) interview educators to gain their insights and perspectives about the tool.
Designing Differentiated Instruction in the Foreign Language Classroom Using ChatGPT
Project Lead: Chiu-Hung Chen, University of Toronto Mississauga, Language Studies
This project aims to explore the possibility of leveraging ChatGPT as a tool to create differentiated instruction for every student or a small group of students who share similar interests and learning profiles, with a special focus on the “what they learn” component in a Chinese language classroom. We will work with teaching assistants and research assistants to:
- Design a survey to evaluate students’ strengths, weaknesses, and personal needs (e.g., communication with grandparents) in a second-year Chinese language course at the beginning of the semester.
- Design and test prompts to help students develop personalized learning activities using ChatGPT. Provide one-on-one or small group coaching to students on creating personalized learning activities using ChatGPT.
- Provide feedback on students’ personalized learning activities.
- Collect feedback from teaching assistants and students on using ChatGPT mid- and end-of-semester.
- Evaluate the feedback and make recommendations.
Exploring Affordances of Generative AI Tools for Pre-service (language) Teacher Education: From lesson development to assessment
Project Lead: Ji-young Shin, University of Toronto Mississauga, Language Studies
This project will explore affordances of generative AI tools for pre-service teacher education, in particular language teachers, based on comprehensive prototypes of language teaching, from teaching material development to classroom teaching, and assessment. The project lead will develop a range of AI-based methods to promote pre-service teachers to understand language learning theories, develop lesson materials, lesson planning, and classroom teaching activities, and finally design classroom assessment throughout the three courses. Based on the identified affordances, the project will propose the prototypes of language learning, teaching, and assessment models using generative AI tools to support (pre-service) teacher education.
Exploring Ethical and Creative Uses of Generative AI to Support Equity-Deserving Students
Project Lead: Elaine Khoo, University of Toronto Scarborough, Centre for Teaching and Learning
Generative AI opens up many opportunities for equity-deserving students in both curricular and co-curricular areas. For the curricular areas, we will develop generative AI usage for English Language Learners (ELLs) and for co-curricular areas, there is immense potential to address students’ language needs in speaking and writing development. We will leverage CTL’s successful Communication Cafes model to incorporate Generative AI that could be used to promote interactional diversity by bringing culturally diverse students together for deep conversation on essential topics that will improve students critical thinking and sense of belonging. The generative AI will be proactively used to provide support from the first day of class according to individual needs so that ELLs initially facing inequitable learning burden would have developed a substantial expansion of their language repertoire to cope with course needs, with a strong foundation of academic integrity values.
Project Leads: Malayna Bernstein, Claire Battershill, Seamus Ross, Matt Ratto; Faculty of Information
Text-based Generative AI systems offer novel capacity for generating quality text (essays, poems, etc.), and much of the current dialogue about AI in educational contexts focuses on preventing academic integrity missteps. In response, our project will shift the dominant policing discourse towards one of opportunity, creativity, and reflection. Generative AI systems open up new ways to learn and offer creative and analytical opportunities for educators. This project will explore these new possibilities by developing a repository for assignments and activities that leverage the affordances of generative AI.
Instructor-AI Collaborative Content Generation Project
Project Leads: Joseph Jay Williams, Nathan Laundry; Faculty of Arts & Science, Computer Science
The development of complex skills like programming is improved by the inclusion of integrative, problem-based learning materials. This project will explore how Large Language Models can be leveraged by instructors in Human-AI collaboration to create textbook materials, example code, and knowledge-testing content for computer science courses efficiently. This work will focus on two goals: exploring the contexts in which LLMs do and do not produce accurate and useful educational materials, and how we can scaffold Instructor-LLM interactions.
This project aims to make the time-consuming process of creating problem-based learning materials for introductory courses faster and easier. Increasing the amount of problem-based learning should improve programming skill development in introductory CS courses.
Intelligent Support for Enhancing Problem Solving, Learning, and Metacognitive Skills in CS Courses: A Novel Tool Combining Large Language Models and Reinforcement Learning
Project Lead: Michael Liut, University of Toronto Mississauga, Mathematical and Computation Sciences
Our project will provide intelligent support specific to students while they solve problems – replicating human tutor actions: suggesting problems, giving explanations, and asking questions. We expand on our work with adaptive personalized explanations, using reinforcement learning (RL), to integrate large language models (LLMs) to provide conversational support. Thus, allowing us to achieve personalized coaching by including content recommendation, metacognitive and motivational support, as well as learner-centered dialogue to enhance the learning experience and objectives for students. This will be done by:
- Enhancing problem-solving abilities within guided authentic tasks by introducing new methods of personalized coaching that leverages LLMs.
- Developing a software tool that provides intelligent and personalized support to students while they solve problems, which can be generalizable to any course.
Learning to Teach and Teaching to Learn in the Generative AI Landscape
Project Leads: Dan Guadagnolo, Sarah Cherki El-Idrissi, Kate Maddalena, Michael Nixon, Steve Szigeti; University of Toronto Mississauga, Institute of Communication, Culture, Information and Technology
This project will support the design and implementation of five distinct but interconnected modules on generative AI across ICCIT’s first-year courses. Modules include:
- Programming assessments to interrogate and troubleshoot AI outputs;
- Primary research activities to responsibly ‘co-write’ with the services;
- Management simulations and decision making tools that capture the complexities of the technology in the information economy;
- Citational analysis that exposes the difficulties LLMs have with referencing; and
- Activities that examine the datasets, labour, & resource costs involved in mounting generative AI technologies.
Each module will share thematic and conceptual overlap, while taking a format specific to the discipline that the faculty lead come from: management information systems and sustainability (Cherki El-Idrissi), computer science (Nixon), information sciences (Szigeti), rhetoric (Maddalena), and media studies (Guadagnolo).
Strengthening Co-op Student’s Communication Skills for Job Search and Workplace Success
Project Leads: Lynn Tucker, Susan Soikie, Cynthia Jairam-Persaud; University of Toronto Scarborough, Arts & Science Co-op
Throughout the Arts & Science Co-op program, students must draw upon their communication skills to successfully progress from the job application, to interview and offer stages, and to excel in the workplace. The department introduced INStage to support practice and supplement traditional assessment. INStage allows students to conduct virtual simulations with a panel of avatars and receive immediate feedback on their communication performance.
With this project, A&S Co-op seeks to enhance the use of this AI- based platform with all co-op students with a particular focus on two student population groups that face barriers to employment – international students and students with disabilities. The expansion of usage with customized support to these target groups will provide students a space to:
- Speak about their unique knowledge and experiences
- Build confidence in communication skills
- Practice communication techniques in Canadian and professional contexts
- Provide students with disabilities a space to practice disclosure and/or requesting accommodations in the workplace.
The project aims to improve student efficacy and confidence in communicating with employers and speaking about themselves and their experiences.
Understanding the Limits of AI-Based Image Generators with DALL·E 2 and Midjourney
Project Lead: Alec Jacobson, Faculty of Arts & Science, Computer Science
In this project, we will create a new assignment sequence for third-year undergraduate students in our Computer Graphics course on the topic of Generative AI. OpenAI’s DALL·E 2 and Midjourney are widely considered state of the art for text to-image generation tools. The success of these models is due in part to their proprietarily large scale. Students will learn how these models work via firsthand experience through pressing these models in directions where they work best and then pressing them into directions that reveal their (possibly negative) biases. Finally, we will explore methods for controlling results in terms of mitigating biases and meeting a user’s artistic goals.
This project is poised for strong student impact by giving a curated experience to generative AI tools that are often out of reach (due to costs) and poorly understood (due to misleading hype in popular media).
The timeline for the LEAF+AI initiative is as follows:
- Projects began in May 2023 and are currently underway.
- Thematic special interest groups have been formed and will meet over the summer.
- Prototype of pilot projects to be completed by end of fall term 2023.
- Project leads to submit a short report on their project in January 2024.
- Community showcase event in winter 2024 followed by the publication of LEAF+AI projects on the website.
Follow this website for updates.
Contact digital.learning@utoronto.ca for more information.