#artificial intelligence
Feb 2025
Mem2Ego: Empowering Vision-Language Models with Global-to-Ego Memory for Long-Horizon Embodied Navigation
A novel vision-language model (VLM)-based navigation framework that addresses challenges by adaptively retrieving task-relevant cues from a global memory module and integrating them with the agent's egocentric observations and enhances spatial reasoning and decision-making in long-horizon tasks.
Ling-Feng Zhang, Yuecheng Liu, Zhan-Guang Zhang et al.
· arXiv.org · 34 citations
· ⚡1