A Hybrid Mamba for Audio-Visual Navigation
For researchers in embodied AI and audio-visual navigation, this work modernizes the backbone architecture that has remained unchanged for five years, offering a more efficient and robust approach.
The paper proposes Samba, a hybrid Mamba architecture for audio-visual navigation that replaces conventional GRUs with a Mamba State Encoder and introduces an Audio Mamba Encoder to capture time-frequency dependencies. On Matterport3D, it improves navigation success rate by 11.3% over state-of-the-art models, with larger gains on Replica.
Since the paradigm centered on convolutional neural networks and recurrent architectures was established in 2020, the fundamental backbone networks for audio-visual navigation have undergone no essential changes for more than five years, making them inadequate to support efficient representation of dynamic multimodal sequences. This paper proposes Samba(A Hybrid Mamba for Audio-Visual Navigation). It uses the adaptive selection-enabled Mamba State Encoder (M-SE) to replace conventional GRUs for temporal aggregation, and constructs an Audio Mamba Encoder (AME) to remedy the limitations of convolutional operators in capturing global time-frequency dependencies in spectrograms. Experiments demonstrate that Samba exhibits exceptional generalization performance when facing unheard sound sources and unseen scenes. On the Matterport3D dataset, it improves the navigation success rate (SR) by 11.3\% compared with existing state-of-the-art models, and the performance gain is even more pronounced on the Replica dataset, which features finer scene structures. Such modernized architectural reconstruction unlocks stronger embodied representation capabilities at a lower computational cost, thereby providing a highly robust technical pathway for paradigm evolution in the field of audio-visual navigation.