Think Hierarchically, Act Dynamically: Hierarchical Multi-modal Fusion and Reasoning for Vision-and-Language Navigation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yue, Junrong, Zhang, Yifan, Qin, Chuan, Li, Bo, Lie, Xiaomin, Yu, Xinlei, Zhang, Wenxin, Zhao, Zhendong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915257911345152
author Yue, Junrong
Zhang, Yifan
Qin, Chuan
Li, Bo
Lie, Xiaomin
Yu, Xinlei
Zhang, Wenxin
Zhao, Zhendong
author_facet Yue, Junrong
Zhang, Yifan
Qin, Chuan
Li, Bo
Lie, Xiaomin
Yu, Xinlei
Zhang, Wenxin
Zhao, Zhendong
contents Vision-and-Language Navigation (VLN) aims to enable embodied agents to follow natural language instructions and reach target locations in real-world environments. While prior methods often rely on either global scene representations or object-level features, these approaches are insufficient for capturing the complex interactions across modalities required for accurate navigation. In this paper, we propose a Multi-level Fusion and Reasoning Architecture (MFRA) to enhance the agent's ability to reason over visual observations, language instructions and navigation history. Specifically, MFRA introduces a hierarchical fusion mechanism that aggregates multi-level features-ranging from low-level visual cues to high-level semantic concepts-across multiple modalities. We further design a reasoning module that leverages fused representations to infer navigation actions through instruction-guided attention and dynamic context integration. By selectively capturing and combining relevant visual, linguistic, and temporal signals, MFRA improves decision-making accuracy in complex navigation scenarios. Extensive experiments on benchmark VLN datasets including REVERIE, R2R, and SOON demonstrate that MFRA achieves superior performance compared to state-of-the-art methods, validating the effectiveness of multi-level modal fusion for embodied navigation.
format Preprint
id arxiv_https___arxiv_org_abs_2504_16516
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Think Hierarchically, Act Dynamically: Hierarchical Multi-modal Fusion and Reasoning for Vision-and-Language Navigation
Yue, Junrong
Zhang, Yifan
Qin, Chuan
Li, Bo
Lie, Xiaomin
Yu, Xinlei
Zhang, Wenxin
Zhao, Zhendong
Computer Vision and Pattern Recognition
Artificial Intelligence
Vision-and-Language Navigation (VLN) aims to enable embodied agents to follow natural language instructions and reach target locations in real-world environments. While prior methods often rely on either global scene representations or object-level features, these approaches are insufficient for capturing the complex interactions across modalities required for accurate navigation. In this paper, we propose a Multi-level Fusion and Reasoning Architecture (MFRA) to enhance the agent's ability to reason over visual observations, language instructions and navigation history. Specifically, MFRA introduces a hierarchical fusion mechanism that aggregates multi-level features-ranging from low-level visual cues to high-level semantic concepts-across multiple modalities. We further design a reasoning module that leverages fused representations to infer navigation actions through instruction-guided attention and dynamic context integration. By selectively capturing and combining relevant visual, linguistic, and temporal signals, MFRA improves decision-making accuracy in complex navigation scenarios. Extensive experiments on benchmark VLN datasets including REVERIE, R2R, and SOON demonstrate that MFRA achieves superior performance compared to state-of-the-art methods, validating the effectiveness of multi-level modal fusion for embodied navigation.
title Think Hierarchically, Act Dynamically: Hierarchical Multi-modal Fusion and Reasoning for Vision-and-Language Navigation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2504.16516