CV

Work Experience

Postdoctoral Fellow The University of Hong Kong (HKU), Hong Kong SAR
Sep. 2025 – Present - Lab: Visual AI Lab

• Advisor: Prof. Kai Han
• Research Focus: Trustworthy Machine Learning, Mobile & GUI Agents, Multi-modal Security, and Optimization Theory.

Education

Ph.D. in Software Engineering (Rank: 1/27) Dalian University of Technology (DLUT), China
Sep. 2021 – Jun. 2025 - Advisor: Prof. Risheng Liu (National Excellent Young Scholar)

• Thesis: Research on Bilevel Optimization and its Applications in Vision and Learning.
• Honors: National Scholarship (2024), Outstanding Graduate.

M.S. in Software Engineering (Rank: 2/54) Dalian University of Technology (DLUT), China
Sep. 2019 – Jun. 2021 - Advisor: Prof. Risheng Liu

• Research: Gradient-based Optimization and Meta-learning.

B.S. in Network Engineering (GPA: 3.96/5.0, Top 10%) Dalian University of Technology (DLUT), China
Sep. 2015 – Jun. 2019 - Core Courses: Mathematical Analysis (99), Probability and Statistics (99).

Research Experience

Postdoctoral Research @ HKU Sep. 2025 – Present - Mobile & GUI Agents: Investigating optimization frameworks for autonomous agents to interact efficiently with mobile interfaces.

• Trustworthy Multi-modal Learning: Exploring security vulnerabilities and defense mechanisms in Large Multi-modal Models (LMMs).

Doctoral Research @ DLUT Sep. 2021 – Jun. 2025 - Bilevel Optimization (BLO): Developed “Augmenting Iterative Trajectory” methods and efficient algorithms for high-dimensional tasks (HO, NAS).

• Representative Works: IEEE TPAMI, NeurIPS (Spotlight), ICML.
• Trustworthy Machine Learning: Designed optimization-based schemes for adversarial attacks and defenses.
  • Representative Works: IJCAI, ACM MM.
• Multimedia Optimization: Applied hierarchical learning to super-resolution and image fusion.
  • Representative Works: IEEE TNNLS, PR, ACM MM.

Master’s Research @ DLUT Sep. 2019 – Jun. 2021 - Open-source Contribution: Built a gradient-based Bilevel Optimization toolbox.

• Award: Best Open-Source Project Award (Second Prize) @ ICME.

Honors & Awards

• 2024: National Scholarship for Outstanding Students
• 2022-2024: Excellent Graduate Student Award (Consecutive 3 years)
• 2021: Best Open-Source Project Award (Second Prize), ICME (CCF B)
• 2019, 2021: First-Class Academic Scholarship for Master’s and Ph.D. Programs
• 2018: Second Prize, 3rd National College Students Academic English Vocabulary Competition
• 2016, 2017: First-Class / Second-Class Academic Scholarship
• 2015, 2016: National Inspirational Scholarship (Top 5%)
• 2016: Second Prize, 25th Dalian City University Students Mathematics Contest

Academic Service

Conference Reviewer / PC Member

• 2025: CVPR, ICML, ICLR, AAAI, ACM MM, ICCV
• 2024: NeurIPS, ICLR, ICML, IJCAI, AAAI, ACM MM
• 2023: NeurIPS, ICLR, ICML, ACM MM
• 2022: NeurIPS, ICLR

Skills

• Programming: Python, MATLAB, PyTorch, C/C++
• Research Areas: Bilevel Optimization, Trustworthy AI, Large Model Fine-tuning, Agent Systems.
• Tools: LaTeX, Git, Linux Environment

Publications

Augmenting Iterative Trajectory for Bilevel Optimization: Methodology, Analysis and Extensions

Risheng Liu, Yaohua Liu, Shangzhi Zeng, Jin Zhang. Augmenting Iterative Trajectory for Bilevel Optimization: Methodology, Analysis and Extensions[J]. Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2025.

Advancing Generalized Transfer Attack with Initialization Derived Bilevel Optimization and Dynamic Sequence Truncation

Yaohua Liu, Jiaxin Gao, Xuan Liu, Xianghao Jiao, Xin Fan, Risheng Liu. Advancing Generalized Transfer Attack with Initialization Derived Bilevel Optimization and Dynamic Sequence Truncation[C]. International Joint Conference on Artificial Intelligence (IJCAI), 2024.

Collaborative brightening and amplification of low-light imagery via bi-level adversarial learning

Jiaxin Gao, Yaohua Liu, Ziyu Yue, et al. Collaborative brightening and amplification of low-light imagery via bi-level adversarial learning[J]. Pattern Recognition (PR), 2024, 154: 110558.

A Dual-Stream-Modulated Learning Framework for Illuminating and Super-Resolving Ultra-Dark Images

Jiaxin Gao, Ziyu Yue, Yaohua Liu, et al. A Dual-Stream-Modulated Learning Framework for Illuminating and Super-Resolving Ultra-Dark Images[J]. IEEE Transactions on Neural Networks and Learning Systems (IEEE TNNLS), 2024.

Enhancing images with coupled low-resolution and ultra-dark degradations: a tri-level learning framework

Jiaxin Gao, Yaohua Liu*. Enhancing images with coupled low-resolution and ultra-dark degradations: a tri-level learning framework[C]. ACM International Conference on Multimedia (ACM MM), 2024.

Motion-Scenario Decoupling for Rat-Aware Video Position Prediction: Strategy and Benchmark

Xiaofeng Liu, Jiaxin Gao, Yaohua Liu, et al. Motion-Scenario Decoupling for Rat-Aware Video Position Prediction: Strategy and Benchmark[C]. International Conference on Image and Graphics (ICIG), 2023, Oral.

PEARL: Preprocessing Enhanced Adversarial Robust Learning of Image Deraining for Semantic Segmentation

Xianghao Jiao, Yaohua Liu, Jiaxin Gao, et al. Pearl: Preprocessing enhanced adversarial robust learning of image deraining for semantic segmentation[C]. ACM International Conference on Multimedia (ACM MM), 2023.

Averaged Method of Multipliers for Bi-Level Optimization without Lower-Level Strong Convexity

Risheng Liu, Yaohua Liu, Wei Yao, Shangzhi Zeng, Jin Zhang. Averaged Method of Multipliers for Bi-Level Optimization without Lower-Level Strong Convexity[C]. International Conference on Machine Learning (ICML), 2023.

Triple-level model inferred collaborative network architecture for video deraining

Pan Mu, Zhu Liu, Yaohua Liu, Risheng Liu, Xin Fan. Triple-level model inferred collaborative network architecture for video deraining[J]. IEEE Transactions on Image Processing (IEEE TIP), 2021, 154: 110558.

Towards Gradient-based Bilevel Optimization with Non-convex Followers and Beyonds

Risheng Liu, Yaohua Liu, Shangzhi Zeng, Jin Zhang. Towards Gradient-based Bilevel Optimization with Non-convex Followers and Beyond[C]. Advances in Neural Information Processing Systems (NeurIPS), 2021, Spotlight, Acceptance Rate ≤ 3% .

Boml: A Modularized Bilevel Optimization Library In Python For Meta Learning

Yaohua Liu, Risheng Liu. Boml: A Modularized Bilevel Optimization Library In Python For Meta Learning[C]. IEEE International Conference on Multimedia & Expo Workshops (ICMEW), 2021, Best Open Source Project Awards, Oral.