Room-to-Room (R-2R)

The 'Room-to-Room (R2R)' dataset is a benchmark that contains navigation tasks requiring agents to follow natural-language instructions to navigate through previously unseen environments, used to evaluate the performance of Vision-and-Language Navigation systems.

Papers using Room-to-Room (R-2R) (4)

Learning to Retrieve Navigable Candidates for Efficient Vision-and-Language Navigation2026Following Route Instructions using Large Vision-Language Models: A Comparison between Low-level and Panoramic Action Spaces2025Grounded Entity-Landmark Adaptive Pre-training for Vision-and-Language Navigation2023NavCoT: Boosting LLM-Based Vision-and-Language Navigation via Learning Disentangled Reasoning2024