Where-OPD: Spatially Guided On-Policy Self-Distillation of MLLMs with Synthetic Scenes Paper • 2610.02117 • Published 3 days ago • 14
Boosting Visual Instruction Tuning with Self-Supervised Guidance Paper • 2604.12966 • Published Apr 14 • 11
DIP: Unsupervised Dense In-Context Post-training of Visual Representations Paper • 2506.18463 • Published Jun 23, 2025 • 21