COM2-03-25
651 66737

ztangent.github.io

TAN Zhi Xuan

NUS Presidential Young Professor
Research Scientist, A*STAR Institute for High Performance Computing

  • Ph.D. (Computer Science, Massachusetts Institute of Technology, 2025)
  • S.M. (Electrical Engineering & Computer Science, Massachusetts Institute of Technology, 2022)
  • B.S. (Electrical Engineering & Computer Science, Yale University, 2018)

Tan Zhi Xuan is a Presidential Young Professor in the NUS Department of Computer Science, with a joint appointment at the A*STAR Institute of High Performance Computing (IHPC). Xuan received their Ph.D. in Computer Science from the Massachusetts Institute of Technology, and her B.S. in Electrical Engineering and Computer Science from Yale University. Xuan's research focuses on scaling cooperative intelligence via rational and model-based AI, spanning the areas of probabilistic programming, model-based planning, Bayesian inference, AI alignment, and computational cognitive science. Together with her research group, the Cooperative Systems & Intelligence (CoSI) lab, Xuan aims to reverse engineer the computational foundations of human cooperation and normativity, thereby enabling the development of AI systems that can cooperate as proficiently as humans, and the design of cooperative sociotechnical infrastructure for an increasingly automated future. Xuan was a recipient of the Open Philanthropy AI Fellowship in 2021 and the A*STAR National Science Scholarship in 2013. In recognition of their expertise in AI safety and alignment, Xuan was invited to participate in the RAISE.SG workshop on Singapore's National AI Strategy 2.0, and has given invited talks at the Simons Institute and the EAGx conference series. She also serves as an advisor to multiple AI alignment non-profits (Principles of Intelligent Behaviour in Biological & Social Systems; Meaning Alignment Institute) and as a board member of Welfare Matters, a non-profit addressing farmed animal welfare in Southeast Asia.

RESEARCH AREAS

Artificial Intelligence
  • Decision Making & Planning
  • Multi-Agent Systems & Algorithmic Game Theory
  • Trustworthy AI

RESEARCH INTERESTS

  • Cooperative AI

  • Bayesian Modeling & Inference

  • Computational Cognitive Science

  • Probabilistic Programming Languages

  • AI Safety & Alignment

RESEARCH PROJECTS

RESEARCH GROUPS

Cooperative Systems & Intelligence (CoSI) Lab

We are dedicated to scaling cooperative intelligence via rational, model-based AI engineering. By reverse engineering the computational foundations of human cooperation, and using those insights to build reliable, coherent, and human-like cooperative systems, we aim to enable collective safety and flourishing in an increasingly automated future.


TEACHING INNOVATIONS

SELECTED PUBLICATIONS

  • Pragmatic Instruction Following and Goal Assistance via Cooperative Language-Guided Inverse Planning
  • Online Bayesian Goal Inference for Boundedly-Rational Planning Agents
  • Learning and Sustaining Shared Normative Systems via Bayesian Rule Induction in Markov Games
  • Sequential Monte Carlo Steering of Large Language Models using Probabilistic Programs
  • Beyond Preferences in AI Alignment

AWARDS & HONOURS

COURSES TAUGHT

CS6208
Advanced Topics in Artificial Intelligence