Zhijing Jin

Zhijing Jin is an Assistant Professor at the University of Toronto and a postdoc at the Max Planck Institute for Intelligent Systems, working with Bernhard Schölkopf. She is also a faculty member at the Vector Institute and an ELLIS advisor. Her research focuses on Causal Reasoning with LLMs, Moral Reasoning in LLMs, and AI Safety, with contributions to AI for Science and NLP for Social Good. She leads the Jinesis AI Lab, which explores causal LLMs, multi-agent systems, and ethical AI applications.

Education:

• PhD in Computer Science from Max Planck Institute (Germany) and ETH Zurich (Switzerland)
• Bachelor’s degree from University of Hong Kong, with visiting semesters at MIT and National Taiwan University

Research Interests:

Her work bridges causal inference and NLP, addressing robustness, interpretability, and ethical alignment of LLMs. Key projects include Corr2Cause (causal reasoning), GovSim/MoralSim (multi-agent LLMs), and frameworks like NLP4SG for social impact. She advocates for causal mechanisms to tackle AI safety and societal challenges.

Recent Articles:

Recent work explores political bias in LLMs, ethical dilemmas in multi-agent systems, and causal foundations for trustworthy AI. These studies emphasize real-world applications, such as healthcare NLP and policy analysis.

Awards & Recognition:

• 3 Rising Star Awards
• 2 Best Paper Awards at NeurIPS 2024 Workshops
• Fellowships from Open Philanthropy and Future of Life Institute

Grants & Mentorship:

Funded by NSERC, Schmidt Sciences, and the Cooperative AI Foundation. She mentors ~20 students globally, including PhD candidates in multi-agent LLMs, causal LLMs, and AI safety. The Jinesis Lab offers remote mentorship across institutions like UofT, ETH Zurich, and University of Michigan.

Labs & Collaborations:

Active collaborations include MPI-IS, Vector Institute, and ETH Zurich’s AI Center. Her lab emphasizes open science, with tools like MoralLens and RouterAttack released publicly.