Cross-modal alignment is one key challenge for Vision-and-Language Navigation (VLN). Most existing studies concentrate on mapping the global instruction or single sub-instruction to the corresponding trajectory. However, another critical problem of achieving fine-grained alignment at the entity level is seldom considered. To address this problem, we propose a novel Grounded Entity-Landmark Adaptive (GELA) pre-training paradigm for VLN tasks. To achieve the adaptive pre-training paradigm, we first introduce grounded entity-landmark human annotations into the Room-to-Room (R2R) dataset, named GEL-R2R. Additionally, we adopt three grounded entity-landmark adaptive pre-training objectives: 1) entity phrase prediction, 2) landmark bounding box prediction, and 3) entity-landmark semantic alignment, which explicitly supervise the learning of fine-grained cross-modal alignment between entity phrases and environment landmarks. Finally, we validate our model on two downstream benchmarks: VLN with descriptive instructions (R2R) and dialogue instructions (CVDN). The comprehensive experiments show that our GELA model achieves state-of-the-art results on both tasks, demonstrating its effectiveness and generalizability.

翻译：跨模态对齐是视觉与语言导航（VLN）的关键挑战之一。现有研究大多集中于将全局指令或单一子指令映射至对应轨迹，然而，实现实体级细粒度对齐的另一关键问题鲜少被关注。为解决此问题，我们提出了一种新颖的具身实体-地标自适应（GELA）预训练范式用于VLN任务。为实现自适应预训练范式，我们首先在Room-to-Room（R2R）数据集中引入具身实体-地标人工标注，命名为GEL-R2R。此外，我们采用三种具身实体-地标自适应预训练目标：1）实体短语预测、2）地标边界框预测、3）实体-地标语义对齐，显式监督实体短语与环境地标之间细粒度跨模态对齐的学习。最终，我们在两个下游基准任务上验证模型：描述性指令导航（R2R）与对话指令导航（CVDN）。全面实验表明，我们的GELA模型在两项任务上均达到最优结果，展示了其有效性与泛化能力。