About Me

Short Bio: I am currently a third-year Ph.D. student in the Research School of Computing, Australian National University, supervised by Prof. Stephen Gould and Prof. Liang Zheng. Before that, I received my master’s degree in computing from Australian National University in 2022 and my bachelor’s degree in mathematics from Peking University in 2019.

Research interest: I have broad interest in computer vision especially autoregressive modeling, including both autoregressive visual understanding and generation.

Publications

Visual Generation

SimFlow: Simplified and End-to-End Training of Latent Normalizing Flows
Qinyu Zhao, Guangting Zheng, Tao Yang, Rui Zhu, Xingjian Leng, Stephen Gould, Liang Zheng
ArXiv preprint, 2025
📄 Preprint | 🌐 Project page | 💻 Code

DiSA: Diffusion Step Annealing in Autoregressive Image Generation
Qinyu Zhao, Jaskirat Singh, Ming Xu, Akshay Asthana, Stephen Gould, Liang Zheng
ArXiv preprint, 2025
📄 Preprint | 🌐 Project page | 💻 Code

ARINAR: Bi-Level Autoregressive Feature-by-Feature Generative Models
Qinyu Zhao, Stephen Gould, Liang Zheng
ArXiv preprint, 2025
📄 Preprint | 💻 Code

Visual Understanding

Can We Predict Performance of Large Models across Vision-Language Tasks?
Qinyu Zhao, Ming Xu, Kartik Gupta, Akshay Asthana, Liang Zheng, Stephen Gould
International Conference on Machine Learning (ICML), 2025
📄 Preprint | 💻 Code

The First to Know: How Token Distributions Reveal Hidden Knowledge in Large Vision-Language Models?
Qinyu Zhao, Ming Xu, Kartik Gupta, Akshay Asthana, Liang Zheng, Stephen Gould
European Conference on Computer Vision (ECCV), 2024
📄 Preprint | 💻 Code

Out-of-Distribution Detection

Towards Optimal Feature-Shaping Methods for Out-of-Distribution Detection
Qinyu Zhao, Ming Xu, Kartik Gupta, Akshay Asthana, Liang Zheng, Stephen Gould
International Conference on Learning Representations (ICLR), 2024
📄 Preprint | 💻 Code

Experience

• ByteDance Seed, China
  • Role: Research Intern in Generative Models
  • Time: Jun 2025 - Dec 2025
• Australian National University, Australia
  • Role: Teaching Assistant
  • Time: Jul 2022 - Present
• Australian National University, Australia
  • Role: Guest Lecturer - Introduce Weights & Biases
  • Time: Sep 2024
  • Description: Advanced Topics in Deep Learning for Computer Vision, 1 lecture

Awards

• Chinese Government Award for Outstanding Self-Financed Students Abroad
  • Description: Granted USD6,000. The highest government award to Chinese doctoral students who study overseas.
  • Time: 2025
• Google Cloud Research Credits Program
  • Description: Granted USD7,808 in Google Cloud Platform credits to support large-scale training of vision-language models.
  • Time: Jul 2024 - Jul 2025
• OpenAI Researcher Access Program
  • Description: Awarded USD5,000 in OpenAI API credits to support research on large multimodal models.
  • Time: Jun 2024 - Dec 2024