About Me

I am a first-year Ph.D. student at HKUST(GZ), AI Thrust, supervised by Prof. Hui Xiong and Prof. Ying Sun.

Advised by Ying Sun and Hui Xiong.

Research Interests

  • AI for ScienceLeveraging advanced AI techniques to empower fundamental scientific research.
  • LLM ReasoningEnhancing the reasoning capacity of large language models through additional context and step-by-step decomposition frameworks.
  • InterpretabilityUnderstanding how LLMs perceive and reason about data.

News

  1. LAYERDOM: An LLM-Emitted Intermediate Representation for Editable Text-in-Image Generation was submitted to AAAI 2027.

  2. Topology-Aware Framework for Chemically Consistent 3D Molecule Generation was submitted to KDD AI4S Track 2027.

  3. PHF: Privileged Hidden Flow for On-Policy Self-Distillation was submitted to AAAI 2027 and released on arXiv.

Publications

  2. Under Review

    LAYERDOM: An LLM-Emitted Intermediate Representation for Editable Text-in-Image Generation

    Lujundong Li, Mingxu Zhang, Songning Lai, Jiayu Yang, Yuxuan Fan, Ziwei Xie, Feng Liu, Changwang Zhang, Jun Wang, Dazhong Shen, Ying Sun

    Submitted to AAAI 2027

    arXivCodeBibTeX
  3. Under Review

    Topology-Aware Framework for Chemically Consistent 3D Molecule Generation

    Lujundong Li, Mingxu Zhang, Dazhong Shen, Ying Sun

    Submitted to KDD AI4S Track 2027

    arXivCodeBibTeX
  11. Under Review

    From Molecular Language to Chemical Cognition: A Survey on Modality Alignment and Knowledge Acquisition in Large Language Models

    Mingxu Zhang*, Yuhan Li*, Dazhong Shen, Ying Sun

    In submission

    arXivCodeBibTeX

Equal contribution.

Experience

  1. –

    Nankai University

    VCIP · Supervised by Mingming Cheng, Xialei Liu

    • Focused on reinforcement learning safety and related algorithmic methods.
    • Collaborated on research projects and paper writing, with an emphasis on empirical evaluation and reproducibility.

  2. –

    Peking University

    PAIR Lab · Supervised by Yaodong Yang

    • Worked on large models, generative models, and safety/alignment-related topics.
    • Conducted literature review and exploration on scalable oversight and weak-to-strong generalization.

Education

  1. –

    Nankai University

    B.Eng. in Computer Science and Engineering

Awards & Honors

  1. HKUST(GZ) PGS Award

  2. Nankai University Excellent Teaching Assistant

  3. China Undergraduate Mathematical Contest in Modeling (Tianjin Area) — Second Prize

  4. Nankai University Academic Excellence Scholarship

  5. Nankai University Innovation Scholarship

  6. Asia and Pacific Mathematical Contest in Modeling — Third Prize (Team Leader)

Academic Service

  1. Artifact Reviewer · KDD 2026