Xin Liu
Higher Degree by Research Candidate
School of Computer Science and Information Technology
College of Engineering and Information Technology
I am a first-year PhD student in Computer and Information Science. Previously, I obtained my Master's degrees in Artificial Intelligence from Southeast University and Monash University.
My research interest is Causal AI. Currently, I focus on integrating foundation models for causal effect estimation. My long-term goal is to develop causality-based methods for trustworthy AI and to leverage AI for scientific discovery. I am also broadly interested in understanding human intelligence and building human-level intelligent systems.
I am broadly interested in the following topics:
- Foundation Models: Tabular foundation models and large language models
- Causal Inference: Causal effect estimation and causal discovery in the era of foundation models
- Causal Representation Learning: Learning causal variables and their relationships from high-dimensional observations
| Language | Competency |
|---|---|
| Chinese (Mandarin) | Can read, write, speak, understand spoken and peer review |
| English | Can read, write, speak, understand spoken and peer review |
| Date | Institution name | Country | Title |
|---|---|---|---|
| 2022 - 2025 | Southeast University | China | Master |
| 2022 - 2025 | Monash University | Australia | Master |
| Year | Citation |
|---|---|
| 2025 | Liu, X., Weijia, Z., & Min-Ling, Z. (2025). HACSurv: A Hierarchical Copula-Based Approach for Survival Analysis with Dependent Competing Risks. In Proceedings of Machine Learning Research. Splash Beach Resort in Mai Khao, Thailand. |
| 2025 | Li, Y., Yang, Z., Huang, Y., Liu, X., & Qi, G. (2025). C^3TG: Conflict-aware, Composite, and Collaborative Controlled Text Generation. In Proceedings of the AAAI Conference on Artificial Intelligence. Singapore. |
| Year | Citation |
|---|---|
| 2025 | Liu, X., Zhang, W., Tang, W., Le, T. D., Li, J., Liu, L., & Zhang, M. -L. (2025). From Correlation to Causation: Max-Pooling-Based Multi-Instance Learning Leads to More Robust Whole Slide Image Classification. |
| 2025 | Li, Y., Yang, Z., Huang, Y., Liu, X., & Qi, G. (2025). C^3TG: Conflict-aware, Composite, and Collaborative Controlled Text Generation. |
Connect With Me
External Profiles
