Multimodal Text Style Transfer for Outdoor Vision-and-Language Navigation
This addresses the problem of limited human-annotated instructions for outdoor navigation agents, though it is incremental as it builds on existing methods to enhance data.
The paper tackles data scarcity in outdoor vision-and-language navigation by introducing a Multimodal Text Style Transfer approach that enriches navigation data using external resources, resulting in an 8.7% relative improvement in task completion rate on the test set.
One of the most challenging topics in Natural Language Processing (NLP) is visually-grounded language understanding and reasoning. Outdoor vision-and-language navigation (VLN) is such a task where an agent follows natural language instructions and navigates a real-life urban environment. Due to the lack of human-annotated instructions that illustrate intricate urban scenes, outdoor VLN remains a challenging task to solve. This paper introduces a Multimodal Text Style Transfer (MTST) learning approach and leverages external multimodal resources to mitigate data scarcity in outdoor navigation tasks. We first enrich the navigation data by transferring the style of the instructions generated by Google Maps API, then pre-train the navigator with the augmented external outdoor navigation dataset. Experimental results show that our MTST learning approach is model-agnostic, and our MTST approach significantly outperforms the baseline models on the outdoor VLN task, improving task completion rate by 8.7% relatively on the test set.