Abstract
Object-goal navigation requires an embodied agent to locate and reach an instance of a specified object category in an indoor environment. Recent training-free approaches leverage vision-language models (VLMs) for open-vocabulary semantic reasoning, but are typically evaluated under an episodic protocol that resets all scene-specific state after each episode. We introduce Cross-Episode Object-Goal Navigation, in which an agent repeatedly operates in the same scene, retains only self-acquired experience, and keeps its model parameters fixed.
To support experience reuse, we present VTM, a training-free VLM navigation framework with a persistent hierarchical Visual-Topological Memory. The VTM organizes scene knowledge at room and object levels and retrieves relevant experience through coarse-to-fine matching, providing memory as soft guidance only when it agrees with current observations. A conservative execution guard further mitigates oscillations, blocked motions, and premature stopping.
Under a controlled same-scene protocol, we evaluate VTM-Nav on three benchmarks: HM3D v0.1, HM3D v0.2, and MP3D, comparing it with a strengthened WMNav baseline augmented with cross-episode textual memory, while keeping the VLM backbone and action pipeline identical. VTM-Nav achieves the best performance across all three benchmarks, demonstrating the effectiveness and robustness of structured visual-topological experience reuse across datasets.
Blogger's Review: The introduction of VTM-Nav marks a significant advancement in the field of goal navigation. The innovative visual-topological memory mechanism not only enhances the model's adaptability but also significantly improves navigation capabilities in complex environments. This success may drive further research into intelligent agents in dynamic settings.