Justin Cui

UofT EngSci + AI Researcher

pfp_green.png

University of Toronto

Toronto, ON, Canada

I am a fourth-year Engineering Science (Machine Intelligence) student at the University of Toronto. My research focuses on human-AI interaction.

I’m interested in building intelligent, fluid interfaces that augment human workflow rather than replace it—supporting complex and creative tasks through systems that enhance rather than automate human capabilities.

Currently, I’m working with Professor Tovi Grossman on LOOM, a proactive learning system that builds personalized knowledge graphs from user interactions. We are continuing to develop visualizations that help users see how concepts connect and identify gaps in their understanding.

Previously, I worked with Professor Scott Sanner on retrieval-augmented conversational recommendation (SIGIR 2024), grounding LLM recommendations in structured data and user reviews. I also explored causal reasoning in LLMs with Professor Zhijing Jin at the Vector Institute and score-based generative models with Professor Alán Aspuru-Guzik.

Reach out at justin.cui@mail.utoronto.ca!

selected publications

  1. AAAI
    LOOM.png
    LOOM: Personalized Learning Informed by Daily LLM Conversations Toward Long-Term Mastery via a Dynamic Learner Memory Graph
    J. Cui, K. Pu, and T. Grossman
    In PerFM @ AAAI, 2026
  2. SIGIR
    SIGIR.png
    Retrieval-Augmented Conversational Recommendation with Prompt-based Semi-Structured Natural Language State Tracking
    S. Kemper*, J. Cui*, K. Dicarlantonio*, K. Lin*, D. Tang*, A. Korikov, and S. Sanner
    * denotes equal contribution
    In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2024
  3. arXiv
    EQR.png
    A Simple but Effective Elaborative Query Reformulation Approach for Natural Language Recommendation
    Q. Wen*, Y. Liu*, J. Cui*, J. Zhang, A. Korikov, G. K. Saad, and S. Sanner
    * denotes equal contribution
    arXiv preprint arXiv:2510.02656, 2025