Advancing General Multimodal Capability of Vision-language Models with Pyramid-descent Visual Position Encoding

Add code
Jan 19, 2025

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: