Can I read Paper 8 on EtoBox?

Paper 8 by Goal Tactics is a document available to read on EtoBox.

What is Paper 8 about?

Uni-NaVid is a video-based Vision-Language-Action model designed to unify various embodied navigation tasks, leveraging 3.6 million navigation samples for training. It processes RGB video streams and natural language instructions to generate low-level robotic actions efficiently, achieving state-of-the-art performance across four navigation tasks: vision-and-language navigation, object-goal navigation, embodied question answering, and human following. The model incorporates an online token merging mechanism

Author

Goal Tactics

Language

EN