arxiv:2607.21556
William Li PRO
williamium
AI & ML interests
Learning AI efficiently and robustly in interactive, multi-modal and embodied environment.
Recent Activity
authored a paper about 8 hours ago
Visual Contrastive Self-Distillation authored a paper about 8 hours ago
From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models authored a paper about 8 hours ago
SPACENUM: Revisiting Spatial Numerical Understanding in VLMs