Xiao-Yu Guo

AI Researcher · PhD in Artificial Intelligence, Monash University

I am an AI researcher with experience across academia and industry. My work focuses on natural language processing, large language models, complex reasoning, multimodal learning, and responsible AI. I am particularly interested in building trustworthy, efficient, and generalisable AI systems.

Previously, I worked as a Senior Applied Scientist at Oracle Australia (January 2025–May 2026), developing medical-domain agent workflows, scalable AI solutions, and methods for model evaluation and validation. Before that, I was a Postdoctoral Research Fellow at the Australian Institute for Machine Learning (AIML) (November 2023–January 2025), working with Ehsan Abbasnejad on model robustness, reliability, and responsible AI.

I completed my PhD in Artificial Intelligence at Monash University in April 2024, under the supervision of Yuan-Fang Li and Gholamreza Haffari. My thesis, Complex Question Answering: from Data to Method, explored reasoning with neural module networks and language models under data bias and distribution challenges.

Xiao-Yu Guo

Research Interests

Natural language processing and large language models; complex question answering and reasoning; multimodal learning; model robustness, fairness, and generalisation; responsible AI.

Selected Publications

  1. Multi-hypothesis Test-Time Adaptation to Mitigate Underspecification
    Afshar Shamsi, Xiao-Yu Guo, Hamid Alinejad-Rokny, Arash Mohammadi, Damien Teney, and Ehsan Abbasnejad
    In Computer Vision – ECCV 2026, 2026
  2. Bayesian Low-Rank Learning (Bella): A Practical Approach to Bayesian Neural Networks
    Bao Gia Doan, Afshar Shamsi, Xiao-Yu Guo, Arash Mohammadi, Hamid Alinejad-Rokny, Dino Sejdinovic, Damien Teney, Damith C. Ranasinghe, and Ehsan Abbasnejad
    In Proceedings of the AAAI Conference on Artificial Intelligence, 2025
  3. Mitigating Language Bias of LMMs in Social Intelligence Understanding with Virtual Counterfactual Calibration
    Peng Chen, Xiao-Yu Guo, Yuan-Fang Li, Xiaowang Zhang, and Zhiyong Feng
    In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024
  4. An Empirical Analysis on Spatial Reasoning Capabilities of Large Multimodal Models
    Fatemeh Shiri, Xiao-Yu Guo, Mona Golestan Far, Xin Yu, Reza Haf, and Yuan-Fang Li
    In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024