Dannong Xu

Doctoral Student

About me:

Hi, I’m Dannong Xu, a Ph.D. student at INSAIT – starting in Jan 2026, under the supervision of Prof. Luc Van Gool and Dr. Danda Paudel. My research interests include Multimodal Learning, Large Language Models, and Computer Vision. Before joining INSAIT, I obtained my Bachelor’s degree from The University of Sydney.

Education:

  • The University of Sydney – Bachelor of Mechanical Engineering (Honours) – Aug 2020 to Feb 2025

Publications:

  • “Script: Graph-Structured and Query-Conditioned Semantic Token Pruning for Multimodal Large Language Models”. TMLR 2025.
  • “WikiAutoGen: Towards Multi-Modal Wikipedia-Style Article Generation”. ICCV 2025.
  • “Document Haystacks: Vision-Language Reasoning Over Piles of 1000+ Documents”. CVPR 2025.

Relevant Experience:

  • EverMind.ai – Long-term Memory
  • KAUST – Multimodal Learning and Computer Vision