Three-Step Nav: A Hierarchical Global-Local Planner for Zero-Shot Vision-and-Language Navigation
Wanrong Zheng, Yunhao Ge, Laurent Itti
TLDR
Three-Step Nav improves zero-shot vision-and-language navigation using a hierarchical global-local planner with a novel three-view protocol.
Key contributions
- Proposes Three-Step Nav, a hierarchical global-local planner for zero-shot VLN.
- Uses a 'look forward' step to extract global landmarks and sketch coarse plans.
- Employs 'look now' for fine-grained guidance and 'look backward' to correct drift.
- Achieves state-of-the-art zero-shot performance on R2R-CE and RxR-CE datasets.
Why it matters
Current MLLM-powered VLN agents struggle with drift and low success rates. Three-Step Nav addresses these by introducing a novel three-view planning protocol. This method significantly improves zero-shot navigation without requiring fine-tuning.
Original Abstract
Breakthrough progress in vision-based navigation through unknown environments has been achieved by using multimodal large language models (MLLMs). These models can plan a sequence of motions by evaluating the current view at each time step against the task and goal given to the agent. However, current zero-shot Vision-and-Language Navigation (VLN) agents powered by MLLMs still tend to drift off course, halt prematurely, and achieve low overall success rates. We propose Three-Step Nav to counteract these failures with a three-view protocol: First, "look forward" to extract global landmarks and sketch a coarse plan. Then, "look now" to align the current visual observation with the next sub-goal for fine-grained guidance. Finally, "look backward" audits the entire trajectory to correct accumulated drift before stopping. Requiring no gradient updates or task-specific fine-tuning, our planner drops into existing VLN pipelines with minimal overhead. Three-Step Nav achieves state-of-the-art zero-shot performance on the R2R-CE and RxR-CE dataset. Our code is available at https://github.com/ZoeyZheng0/3-step-Nav.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.