场论记忆用于人工智能代理:连续动态以保持上下文
Field-Theoretic Memory for AI Agents: Continuous Dynamics for Context Preservation
AI总结:
本文提出基于场论的记忆系统,通过连续动态保持上下文,提升多会话推理和时间推理性能。
摘要:
我们提出了一种用于人工智能代理的记忆系统,将存储的信息视为由偏微分方程支配的连续场,而不是数据库中的离散条目。该方法借鉴了经典场论:记忆通过语义空间扩散,根据重要性热力学衰减,并在多代理场景中通过场耦合相互作用。我们评估该系统在两个已建立的长上下文基准上的表现:LoCoMo(ACL 2024)具有300轮对话跨越35个会话,以及LongMemEval(ICLR 2025)测试多会话推理超过500轮。在LongMemEval上,场论方法取得了显著改进:在多会话推理上F1提高116%(p<0.01,d=3.06),在时间推理上提高43.8%(p<0.001,d=9.21),在知识更新检索召回上提高27.8%(p<0.001,d=5.00)。多代理实验显示通过场耦合实现了近完美的集体智能(>99.8%)。代码可在github.com/rotalabs/rotalabs-fieldmem获取。
英文摘要:
We present a memory system for AI agents that treats stored information as continuous fields governed by partial differential equations rather than discrete entries in a database. The approach draws from classical field theory: memories diffuse through semantic space, decay thermodynamically based on importance, and interact through field coupling in multi-agent scenarios. We evaluate the system on two established long-context benchmarks: LoCoMo (ACL 2024) with 300-turn conversations across 35 sessions, and LongMemEval (ICLR 2025) testing multi-session reasoning over 500+ turns. On LongMemEval, the field-theoretic approach achieves significant improvements: +116% F1 on multi-session reasoning (p<0.01, d= 3.06), +43.8% on temporal reasoning (p<0.001, d= 9.21), and +27.8% retrieval recall on knowledge updates (p<0.001, d= 5.00). Multi-agent experiments show near-perfect collective intelligence (>99.8%) through field coupling. Code is available at github.com/rotalabs/rotalabs-fieldmem.
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。