CraniMem: Cranial Inspired Gated and Bounded Memory for Agentic Systems
CraniMem:受大脑启发的门控和有界的记忆用于智能体系统
机构 * Dwarkadas Jivanlal Sanghvi College of Engineering(达沃拉萨·吉文拉尔·桑格维工程学院)
专题命中 记忆与上下文管理 :agentic(title,abstract);agent(abstract);分类 cs.AI
AI总结 CraniMem通过门控和有界多阶段记忆设计提升智能体在长期任务中的稳定性与鲁棒性,采用目标条件门控和效用标签结合短期缓冲和长期知识图谱,实现更持久的语义记忆与更小的干扰影响。
Comments International Conference on Learning Representations 2026 Workshop on Memory for LLM-Based Agentic Systems (MemAgents)