AI & ML Paradigm Shift

REST transforms the zero-shot object-navigation problem from simple waypoint selection to a tree-of-paths reasoning process.

March 20, 2026

Original Paper

REST: Receding Horizon Explorative Steiner Tree for Zero-Shot Object-Goal Navigation

Shuqi Xiao, Maani Ghaffari, Chengzhong Xu, Hui Kong

arXiv · 2603.18624

The Takeaway

By representing potential navigation options as a tree of paths rather than isolated destinations, LLMs can reason about en-route information gain. This hierarchical search significantly improves efficiency and success rates in unknown environments without any task-specific training.

From the abstract

Zero-shot object-goal navigation (ZSON) requires navigating unknown environments to find a target object without task-specific training. Prior hierarchical training-free solutions invest in scene understanding (\textit{belief}) and high-level decision-making (\textit{policy}), yet overlook the design of \textit{option}, i.e., a subgoal candidate proposed from evolving belief and presented to policy for selection. In practice, options are reduced to isolated waypoints scored independently: single