aiqbench analysis
Graph-MambaNav advances object-goal navigation by integrating Mamba-based selective scanning with LLM-derived commonsense priors to model long-range dependencies in unseen environments. This approach addresses critical limitations in existing graph methods regarding permutation invariance and information propagation order, enabling more efficient reasoning over complex spatial-temporal relationships for robotic agents.
Graph-MambaNav: Spatial-Temporal Graph Mamba Leveraging Object-Relation Knowledge for Object-Goal Navigation
arXiv:2608.13723v1 Announce Type: new Abstract: Object-goal navigation requires an agent to reason over object relationships and prioritize target-relevant objects for efficient decision making in unseen environments. While existing graph-based methods incorporate target-awareness at the feature or attention level, they remain permutation-invariant and lack an explicit mechanism to control information propagation order, limiting their ability to model target-dependent importance and long-range dependencies. In contrast, Graph-Mamba highlights that node prioritization through sequence ordering is critical for effective global reasoning. In this work, we investigate the node prioritization mechanism in Graph-Mamba and study its role in object navigation. We propose Graph-MambaNav, a target-aware spatial-temporal graph encoding framework that introduces a heuristic ordering over objects based on their relevance to the target, allowing more informative objects to be processed later to aggregate richer context. Both node ordering and edge weights are initialized from LLM-derived commonsense object relationships, providing a unified prior for structured reasoning. A spatial module integrates local message passing with global GraphMamba-based selective scanning, while a temporal module applies Mamba-based sequence modeling over object-wise temporal orders, allowing selective aggregation of historical context for long-range temporal reasoning. Experiments on AI2-THOR and RoboTHOR demonstrate improved navigation performance with generalization, and additional real-world robot deployment further validates the effectiveness of our proposed approach.
Abstract:Object-goal navigation requires an agent to reason over object relationships and prioritize target-relevant objects for efficient decision making in unseen environments. While existing graph-based methods incorporate target-awareness at the feature or attention level, they remain permutation-invariant and lack an explicit mechanism to control information propagation order, limiting their ability to model target-dependent importance and long-range dependencies. In contrast, Graph-Mamba highlights that node prioritization through sequence ordering is critical for effective global reasoning. In this work, we investigate the node prioritization mechanism in Graph-Mamba and study its role in object navigation. We propose Graph-MambaNav, a target-aware spatial-temporal graph encoding framework that introduces a heuristic ordering over objects based on their relevance to the target, allowing more informative objects to be processed later to aggregate richer context. Both node ordering and edge weights are initialized from LLM-derived commonsense object relationships, providing a unified prior for structured reasoning. A spatial module integrates local message passing with global GraphMamba-based selective scanning, while a temporal module applies Mamba-based sequence modeling over object-wise temporal orders, allowing selective aggregation of historical context for long-range temporal reasoning. Experiments on AI2-THOR and RoboTHOR demonstrate improved navigation performance with generalization, and additional real-world robot deployment further validates the effectiveness of our proposed approach.
Robotics (cs.RO)
arXiv:2608.13723 [cs.RO]
(or arXiv:2608.13723v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2608.13723 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 296 words
Related coverage
CDKF-Track: Cluster-aware Data-Driven Kalman Filtering for Cooperative 3D Multi-Object Tracking
arXiv · Sep 24, 2026
Graph Mamba Operator: A Latent Simulator for Interacting Particle Systems
arXiv · Jun 9, 2026
TBSG-Net: Temporal Bipartite Scene Graph Network for Fine-Grained Video Moment Retrieval
arXiv · Aug 5, 2026
STEGNav: Spatio-Temporal Event Graph Reasoning for Multimodal Lifelong Object Navigation
arXiv · Aug 31, 2026