Search papers, labs, and topics across Lattice.
This paper introduces VTM-Nav, a hierarchical Visual-Topological Memory framework designed for Cross-Episode Object-Goal Navigation, allowing an agent to effectively reuse self-acquired experience in a fixed scene. By organizing scene knowledge at both room and object levels and employing a coarse-to-fine matching retrieval process, VTM-Nav enhances navigation performance without retraining the model. The results show that VTM-Nav outperforms a baseline method across multiple benchmarks, highlighting the advantages of structured memory in embodied navigation tasks.
Experience reuse in navigation can significantly boost performance, with VTM-Nav outperforming traditional methods by leveraging a hierarchical memory structure.
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 \method, a training-free VLM navigation framework with a persistent hierarchical Visual-Topological Memory (VTM). 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 \method{} on three benchmarks, HM3D v0.1, HM3D v0.2, and MP3D, and compare it with a strengthened WMNav baseline augmented with cross-episode textual memory, while keeping the VLM backbone and action pipeline identical. \method{} achieves the best performance across all three benchmarks, demonstrating the effectiveness and robustness of structured visual-topological experience reuse across datasets.
