AI RESEARCH
Self-Improving Small Object Grounding in LVLMs
arXiv CS.LG
•
ArXi:2606.01612v1 Announce Type: cross Can internal attention patterns in Large Vision Language Models (LVLMs) identify reliable small-object boxes without fine-tuning? In this work, we provide an affirmative answer. Attention structure in LVLMs encodes grounding quality-a lightweight IoU regressor trained solely on attention maps achieves strong IoU prediction (Pearson r > 0.67