RxR

The 'RxR' dataset is a benchmark used to evaluate Vision-Language Navigation (VLN) tasks, containing navigation instructions paired with visual data to assess the performance of models in understanding and executing these instructions.

Papers using RxR (11)

pFedNavi: Structure-Aware Personalized Federated Vision-Language Navigation for Embodied AI2026Fine-Grained Instruction-Guided Graph Reasoning for Vision-and-Language Navigation2025Vision-and-Language Navigation via Causal Learning2024 · 28 citesHistory Aware Multimodal Transformer for Vision-and-Language Navigation2021 · 21 citesHOP: History-and-Order Aware Pre-training for Vision-and-Language Navigation2022 · 8 citesA New Path: Scaling Vision-and-Language Navigation with Synthetic Instructions and Imitation Learning2022 · 1 citesVLN-PETL: Parameter-Efficient Transfer Learning for Vision-and-Language Navigation2023Causality-based Cross-Modal Representation Learning for Vision-and-Language Navigation2024DELAN: Dual-Level Alignment for Vision-and-Language Navigation by Cross-Modal Contrastive Learning2024Why Only Text: Empowering Vision-and-Language Navigation with Multi-modal Prompts2024NAVCON: A Cognitively Inspired and Linguistically Grounded Corpus for Vision and Language Navigation2024