HCMI Lab / Toronto Metropolitan University
We study how people search, trust and govern information in a world shaped by machine intelligence.
Latest News
Openings for PhD, MASc, MEng and undergraduate students, visiting researchers and postdoctoral fellows. Read the application requirements before contacting the lab.
See Openings & How to ApplyAn evidence selection approach for improving LLM-based query expansion.
→Graph neural networks with kernel pooling for finding individual experts and forming collaborative expert teams on community question answering platforms.
→Adversarial content injection attacks that mislead neural ranking models.
→Reasoning over query difficulty for query performance prediction.
→Characterizing moderator disagreement and inconsistency in Reddit content moderation.
→Ranking local vote authority to support platform governance decisions.
→Research
Generative search and LLM-based retrieval; LLM-as-a-judge evaluation; fair and bias-aware ranking; adversarial robustness of neural rankers; query performance prediction; platform governance and decentralized autonomous organizations.
Learn moreIndustry, academic, and funding partners, past and present.
