Full Professor, Mohamed bin Zayed University of Artificial Intelligence
3 papers at NeurIPS 2025
An efficient fine-tuning framework that adapts a pre-trained weight matrix by decomposing its update into two components: (1) a projection onto the complement of a low-rank subspace spanned by a low-rank matrix, and (2) a low-rank update.
A novel method to detect and localise forgeries coming from diffusion-based inpainter models.
We introduce PhyBlock, a progressive benchmark evaluating large vision-language models on physical understanding and spatial planning via robotic 3D block assembly tasks.