2026
FILD-Nav:Vision-and-Language Navigation with Instruction Landmark Features in Continuous Environments
IJCAI 2026
Vision-and-language navigation (VLN) requires agents to follow natural language instructions to navigate autonomously in continuous environments. However, existing approaches often lack high-level semantic guidance in waypoint prediction and explicit language–landmark alignment in cross-modal planni