Fetching the paper…

LLM-Grounder: Open-Vocabulary 3D Visual Grounding with Large Language Model as an Agent · Around