You can find the full technical details on arXiv: VL-Nav .
Uses a CVL (Curiosity-driven Vision-Language) score to prioritize exploring unknown areas that align with human descriptions. Sandris Dubovs V L Nav Neka
"In rigorous testing, including the , VL-Nav achieved a 75–83% success rate across indoor and outdoor settings. In real-world deployments, it maintained an 86.3% success rate , demonstrating reliability over long-range trajectories of up to 483 meters." Resources for Further Development You can find the full technical details on arXiv: VL-Nav
View demonstrations on robots like the Unitree G1 and Go2 at the SAIR Lab Project Page . it maintained an 86.3% success rate
Leverages a 3D scene graph and image memory to help Vision Language Models (VLMs) replan tasks in real-time.