RoboMemory: A Brain-inspired Multi-memory Agentic Framework for Interactive Environmental Learning in Physical Embodied Systems
Brain-inspired memory for embodied agents (temporal, spatial, semantic, episodic) with a spatial knowledge graph. Reports +26.5 points for Qwen2.5-VL-72B on EmbodiedBench.受大脑启发的具身智能体记忆框架（时间、空间、语义、情景记忆），配合动态空间知识图谱。报告让 Qwen2.5-VL-72B 在 EmbodiedBench 上平均成功率提高 26.5 个百分点。
BibTeX
@misc{lei2025robomemory,
title = {{RoboMemory: A Brain-inspired Multi-memory Agentic Framework for Interactive Environmental Learning in Physical Embodied Systems}},
author = {Mingcong Lei and Honghao Cai and Yuyuan Yang and Yimou Wu and Jinke Ren and Zezhou Cui and Liangchen Tan and Junkun Hong and Gehan Hu and Shuangyu Zhu and Shaohan Jiang and Ge Wang and Junyuan Tan and Zhenglin Wan and Zheng Li and Zhen Li and Shuguang Cui and Yiming Zhao and Yatong Han},
year = {2025},
eprint = {2508.01415},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2508.01415},
}