About me

I am a PhD candidate in wis group, TU Delft, supervised by Prof.dr. Ujwal Gadiraju and Prof.dr.ir. Geert-Jan Houben. I obtained my master and bachelor degree from Information School at Renmin University of China (RUC), supervised by Prof.dr. Wayne Xin Zhao and Prof.dr. Ji-Rong Wen. I also worked as a visiting research student in Singapore Management University, supervised by Prof.dr. Jing Jiang.

Research Interest

In my master program, I primarily focused on deep learning techniques for graph representation learning and reasoning, and on intelligent applications with relational knowledge bases, such as knowledge base question answering, knowledge graph completion, and knowledge-based recommender systems.

My current research focuses on human-centered AI:

  • Exploring how Large Language Models (LLMs) can collaborate with humans to boost work efficiency and optimize workflow.
  • Analyzing user factors (e.g., cognitive bias) in shaping trust and reliance in human-AI collaboration.
  • Designing effective explanation methods (e.g., using analogy) to facilitate human-AI collaboration (target users: laypeople, non-experts)
  • Collecting necessary human knowledge (e.g., commonsense knowledge) to improve AI systems and generate explanations

Selected Publication

  • Knowing About Knowing: An Illusion of Human Competence Can Hinder Appropriate Reliance on AI Systems.
  • It Is Like Finding a Polar Bear in the Savannah! Concept-level AI Explanations with Analogical Inference from Commonsense Knowledge.
    • Gaole He, Agathe Balayn, Stefan Buijsman, Jie Yang, Ujwal Gadiraju. HCOMP 2022 (10th AAAI Conference on Human Computation and Crowdsourcing).
    • [paper], [code], [slides], [video], [blog]
    • Best Paper Award.
  • Ready Player One! Eliciting Diverse Knowledge Using A Configurable Game.
    • Agathe Balayn*, Gaole He*, Andrea Hu*, Jie Yang, Ujwal Gadiraju. WWW 2022.
    • [paper], [code], [slides]
    • Best Paper Candidate, honorable nomination.
  • Improving Multi-hop Knowledge Base Question Answering by Learning Intermediate Supervision Signals.
  • Mining Implicit Entity Preference from User-Item Interaction Data for Knowledge Graph Completion via Adversarial Learning.

Selected Awards

  • Gary Marsden Travel Award for CHI’23, 2023
  • HCOMP Best Paper Award, 2022
  • Excellent Graduate in Beijing, 2021
  • WSDM travel grant, 2021
  • JingDong Scholarship, 2020 (Special First-Class scholarship for postgraduate student)
  • American College Mathematical Modeling Contest Meritorious Winner, 2016

Contact

Building 28, Van Mourik Broekmanweg 6, 2628 XE Delft, the Netherlands

Email: g.he@tudelft.nl