arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2604.16331cs.ROcs.AIcs.CVcs.MA

BrainMem: 为具身智能任务规划设计的脑启发式记忆系统

BrainMem: Brain-Inspired Evolving Memory for Embodied Agent Task Planning

Nanyang Technological University, Singapore（南洋理工大学，新加坡）；Tianjin University, Tianjin, China（天津大学，天津，中国）

Xiaoyu Ma, Lianyu Hu, Wenbing Tang, Zixuan Hu, Zeqin Liao, Zhizhen Wu, Yang Liu

阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出BrainMem，一种无训练的分层记忆系统，通过构建知识图谱和符号指南，提升具身智能在复杂环境中的任务规划能力，尤其在长周期和空间复杂任务中表现突出。

摘要：

具身任务规划需要智能体在复杂3D环境中执行长周期、目标导向的动作，其成功依赖即时感知和跨任务积累的经验。然而，现有基于LLM的规划器多为无状态和反应式，缺乏持久记忆，导致重复错误和难以处理空间或时间依赖性。本文提出BrainMem（脑启发式记忆），一种无训练的分层记忆系统，通过将交互历史转化为结构化知识图谱和提炼的符号指南，使智能体能够检索、推理和适应过去经验，而无需模型微调或额外训练。该设计无缝集成任意多模态LLM，大幅减少对任务特定提示工程的依赖。在四个代表性基准测试中，BrainMem显著提升了任务成功率，尤其在长周期和空间复杂任务中效果最明显，展示了进化记忆作为通用具身智能的有前途且可扩展的机制。

英文摘要：

Embodied task planning requires agents to execute long-horizon, goal-directed actions in complex 3D environments, where success depends on both immediate perception and accumulated experience across tasks. However, most existing LLM-based planners are stateless and reactive, operating without persistent memory and therefore repeating errors and struggling with spatial or temporal dependencies. We propose BrainMem(Brain-Inspired Evolving Memory), a training-free hierarchical memory system that equips embodied agents with working, episodic, and semantic memory inspired by human cognition. BrainMem continuously transforms interaction histories into structured knowledge graphs and distilled symbolic guidelines, enabling planners to retrieve, reason over, and adapt behaviors from past experience without any model fine-tuning or additional training. This plug-and-play design integrates seamlessly with arbitrary multi-modal LLMs and greatly reduces reliance on task-specific prompt engineering. Extensive experiments on four representative benchmarks, including EB-ALFRED, EB-Navigation, EB-Manipulation, and EB-Habitat, demonstrate that BrainMem significantly enhances task success rates across diverse models and difficulty subsets, with the largest gains observed on long-horizon and spatially complex tasks. These results highlight evolving memory as a promising and scalable mechanism for generalizable embodied intelligence.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑