MemVLN: Episodic and Procedural Memory for Vision-and-Language Navigation
For VLN-CE agents, this work solves the trade-off between trajectory consistency and inference speed, enabling practical real-time navigation with an LLM-based framework.
MemVLN achieves state-of-the-art performance on VLN-CE benchmarks (5.8% SR improvement on R2R, 9.7% on RxR) while enabling real-time inference at 14 FPS, using episodic and procedural memory to balance long-horizon visual history with low-latency action execution.
Vision-and-Language Navigation in Continuous Environments (VLN-CE) requires agents to maintain long-horizon visual history for trajectory consistency while executing actions with low latency. Existing video-based VLN approaches typically struggle to satisfy both demands simultaneously. To address these challenges, we propose MemVLN, a novel VLN framework that achieves state-of-the-art performance with real-time inference efficiency (14 FPS). MemVLN utilizes a visual encoder to process continuous observations and a Large Language Model (LLM) to interpret instructions and generate actions. Central to our approach is an Episodic Memory management that applies pyramidal resolutions. This mechanism concentrates computation on immediate percepts while retaining compressed long-term history. Complementing to this design, we introduce Procedural Memory for fast action with a compact vocabulary of atomic mid-level actions to bypass auto-regressive decoding latency. Experiments on VLN-CE show that MemVLN-4B surpasses the baseline Qwen3-VL-4B architecture by 5.8\% SR in R2R and 9.7\% SR in RxR, while achieving a 7$\times$ speedup in inference latency.