R-2R-CE

The 'R2R-CE' dataset/benchmark contains a collection of navigation tasks that evaluate an agent's ability to follow natural-language instructions in unseen environments.

Papers using R-2R-CE (6)

P2DNav: Panorama-to-Downview Reasoning for Zero-shot Vision-and-Language Navigation2026Boosting Zero-shot VLN Via Abstract Obstacle Map-based Waypoint Prediction With Topograph-and-visitinfo-aware Prompting2025NavForesee: A Unified Vision-Language World Model for Hierarchical Planning and Dual-Horizon Navigation Prediction2025D3D-VLP: Dynamic 3D Vision-Language-Planning Model for Embodied Grounding and Navigation2025One Agent to Guide Them All: Empowering MLLMs for Vision-and-Language Navigation via Explicit World Representation2026SpatialAnt: Autonomous Zero-Shot Robot Navigation via Active Scene Reconstruction and Visual Anticipation2026