arXiv cs.AI· Xueyao Wei, Yifei Wang, Delun Li, Shaoan Wang, Junzhi Yu, Wenjie Song, Xingyu Chen·· 3 天前AI 评分22
MemoryVLN：从自我感知记忆表示到视觉语言导航中的跨上下文记忆复用
MemoryVLN: From Ego-Aware Memory Representation to Cross-Context Memory Reuse in Vision-Language Navigation
AI 导读
针对大型视觉语言模型在长时程视觉语言导航中因上下文长度限制、历史观测冗余和空间对齐不稳定导致性能下降的问题，研究者提出 MemoryVLN 记忆驱动导航系统，通过长序列压缩、自我感知记忆表示、分层记忆库与记忆增强机制，将原始导航历史转化为紧凑、空间锚定且可选择性检索复用的导航经验。
正文 · 原文
Abstract:Large vision-language models (LVLMs) have shown strong potential for embodied vision-language navigation (VLN), yet their performance degrades on long-horizon trajectories due to context-length limits, redundant historical observations, and unstable spatial alignment across time. These issues make it difficult to both compress long navigation histories and reliably reuse past spatial experience. To address these limitations, we propose \textbf{MemoryVLN}, a unified memory-driven navigation system built around memory construction and management. During memory construction, \emph{Long Sequence Compression} employs forgetting-curve-inspired temporal sampling to preserve dense recent evidence and sparse yet representative distant observations under a limited token budget. \emph{Ego-Aware Memory Representation} further integrates 2D semantic tokens, 3D geometric tokens, and 6D agent poses, allowing each memory to encode both what was observed and where it was observed. During memory management, \emph{Hierarchical Memory Bank} maintains in-episode and context-level memories through cross-context warm-start and confidence-controlled write-back. \emph{Memory Enhancement} then retrieves the Top-$K$ relevant memories according to the current navigation state and fuses them with the current visual tokens. Together, these mechanisms transform raw navigation history into compact, spatially grounded, selectively retrievable, and reusable navigation experience. Extensive experiments on R2R-CE and RxR-CE demonstrate consistent improvements over the state-of-the-art baselines by a substantial margin (i.e., 2\% and 6.5\% in successful rate).
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.18478 [cs.RO] |
| (or arXiv:2610.18478v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.18478 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xueyao Wei [view email]
[v1]
Tue, 13 Oct 2026 14:40:04 UTC (13,706 KB)
来源：arXiv cs.AI · arxiv.org