MapDream: Task-Driven Map Learning for Vision-Language Navigation

• URL: http://arxiv.org/abs/2602.00222v2
• Date: Tue, 03 Feb 2026 09:27:42 GMT
• Title: MapDream: Task-Driven Map Learning for Vision-Language Navigation
• Authors: Guoxin Lian, Shuo Wang, Yucheng Wang, Yongcai Wang, Maiyue Chen, Kaihui Wang, Bo Zhang, Zhizhong Su, Deying Li, Zhaoxin Fan,
• Abstract summary: Vision-Language Navigation (VLN) requires agents to follow natural language instructions in partially observed 3D environments.<n>We propose MapDream, a map-in-the-loop framework that formulates map construction as autoregressive bird's-eye-view (BEV) image synthesis.<n> Experiments on R2R-CE and RxR-CE achieve state-of-the-art monocular performance, validating task-driven generative map learning.
• Score: 24.46479241004971
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision-Language Navigation (VLN) requires agents to follow natural language instructions in partially observed 3D environments, motivating map representations that aggregate spatial context beyond local perception. However, most existing approaches rely on hand-crafted maps constructed independently of the navigation policy. We argue that maps should instead be learned representations shaped directly by navigation objectives rather than exhaustive reconstructions. Based on this insight, we propose MapDream, a map-in-the-loop framework that formulates map construction as autoregressive bird's-eye-view (BEV) image synthesis. The framework jointly learns map generation and action prediction, distilling environmental context into a compact three-channel BEV map that preserves only navigation-critical affordances. Supervised pre-training bootstraps a reliable mapping-to-control interface, while the autoregressive design enables end-to-end joint optimization through reinforcement fine-tuning. Experiments on R2R-CE and RxR-CE achieve state-of-the-art monocular performance, validating task-driven generative map learning.

Related papers

• SkeNa: Learning to Navigate Unseen Environments Based on Abstract Hand-Drawn Maps [20.963573846962987]
  We introduce Sketch map-based visual Navigation (SkeNa)<n>SkeNa is an embodied navigation task in which an agent must reach a goal in an unseen environment using only a hand-drawn sketch map as guidance.<n>We present a large-scale dataset named SoR, comprising 54k trajectory and sketch map pairs across 71 indoor scenes.
  arXiv  Detail & Related papers  (2025-08-05T03:56:32Z)
• Move to Understand a 3D Scene: Bridging Visual Grounding and Exploration for Efficient and Versatile Embodied Navigation [54.04601077224252]
  Embodied scene understanding requires not only comprehending visual-spatial information but also determining where to explore next in the 3D physical world.<n>underlinetextbf3D vision-language learning enables embodied agents to effectively explore and understand their environment.<n>model's versatility enables navigation using diverse input modalities, including categories, language descriptions, and reference images.
  arXiv  Detail & Related papers  (2025-07-05T14:15:52Z)
• Object Goal Navigation with Recursive Implicit Maps [92.6347010295396]
  We propose an implicit spatial map for object goal navigation. Our method significantly outperforms the state of the art on the challenging MP3D dataset. We deploy our model on a real robot and achieve encouraging object goal navigation results in real scenes.
  arXiv  Detail & Related papers  (2023-08-10T14:21:33Z)
• Learning Navigational Visual Representations with Semantic Map Supervision [85.91625020847358]
  We propose a navigational-specific visual representation learning method by contrasting the agent's egocentric views and semantic maps. Ego$2$-Map learning transfers the compact and rich information from a map, such as objects, structure and transition, to the agent's egocentric representations for navigation.
  arXiv  Detail & Related papers  (2023-07-23T14:01:05Z)
• BEVBert: Multimodal Map Pre-training for Language-guided Navigation [75.23388288113817]
  We propose a new map-based pre-training paradigm that is spatial-aware for use in vision-and-language navigation (VLN) We build a local metric map to explicitly aggregate incomplete observations and remove duplicates, while modeling navigation dependency in a global topological map. Based on the hybrid map, we devise a pre-training framework to learn a multimodal map representation, which enhances spatial-aware cross-modal reasoning thereby facilitating the language-guided navigation goal.
  arXiv  Detail & Related papers  (2022-12-08T16:27:54Z)
• Weakly-Supervised Multi-Granularity Map Learning for Vision-and-Language Navigation [87.52136927091712]
  We address a practical yet challenging problem of training robot agents to navigate in an environment following a path described by some language instructions. To achieve accurate and efficient navigation, it is critical to build a map that accurately represents both spatial location and the semantic information of the environment objects. We propose a multi-granularity map, which contains both object fine-grained details (e.g., color, texture) and semantic classes, to represent objects more comprehensively.
  arXiv  Detail & Related papers  (2022-10-14T04:23:27Z)
• Cross-modal Map Learning for Vision and Language Navigation [82.04247028482244]
  We consider the problem of Vision-and-Language Navigation (VLN) In contrast to other works, our key insight is that the association between language and vision is stronger when it occurs in explicit spatial representations. We propose a cross-modal map learning model for vision-and-language navigation that first learns to predict the top-down semantics on an egocentric map for both observed and unobserved regions.
  arXiv  Detail & Related papers  (2022-03-10T03:30:12Z)
• Lifelong Topological Visual Navigation [16.41858724205884]
  We propose a learning-based visual navigation method with graph update strategies that improve lifelong navigation performance over time. We take inspiration from sampling-based planning algorithms to build image-based topological graphs, resulting in sparser graphs yet with higher navigation performance compared to baseline methods. Unlike controllers that learn from fixed training environments, we show that our model can be finetuned using a relatively small dataset from the real-world environment where the robot is deployed.
  arXiv  Detail & Related papers  (2021-10-16T06:16:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.