LifeFuse-Mem: Lifecycle-Aware State Fusion Against Temporary Overwriting for Long-Term Memory

📅 2026-09-11
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
为解决长期记忆中临时信息覆盖持久知识的问题,提出LifeFuse-Mem框架,通过生命周期感知的更新方法分离不同类型的信息。
📝 Abstract
Long-running LLM agents require memory mechanisms that maintain coherent internal states across interactions. We study a lifecycle-labeled memory setting in which write episodes provide lifecycle metadata during training, and phase-aware readout is used during evaluation. This setting reflects the need to distinguish information that should remain influential across future interactions from information that should affect only the current context. A mismatch between these lifecycles can cause temporary information to overwrite durable knowledge, leading to behavioral drift in persistent agents. Within this setting, we introduce \textbf{LifeFuse-Mem}, a lifecycle-aware neural memory framework that separates information according to its temporal commitment. LifeFuse-Mem uses dedicated memory components and lifecycle-aware updates to allow stable and transient knowledge to evolve locally without converting temporary context into durable state. On the controlled anti-overwrite benchmark, LifeFuse-Mem improves acquisition-controlled retention and reduces temporary overwrite; on two public long-memory benchmarks, it remains broadly competitive. These results suggest that explicit lifecycle signals can help diagnose and mitigate overwrite in compact online memory.
Problem

Research questions and friction points this paper is trying to address.

long-running LLM agents
memory mechanisms
lifecycle-aware
temporary overwriting
behavioral drift
Innovation

Methods, ideas, or system contributions that make the work stand out.

Lifecycle-Aware
State Fusion
Temporary Overwriting
Long-Term Memory
Neural Memory Framework
Hanyu Zhao
Hanyu Zhao
Alibaba Group
Distributed SystemsSystems for AI
Y
Yuqian Feng
University of Chinese Academy of Sciences
Z
Zhenyu Song
National University of Defense Technology
Y
Yuanchao Cheng
Beijing Academy of Artificial Intelligence
Y
Yance Jiao
Beijing Academy of Artificial Intelligence
T
Tengfei Pan
Beijing Academy of Artificial Intelligence
L
Li Du
Beijing Academy of Artificial Intelligence