LifeMem improves continual skill reuse for AI agents in many tasks
LifeMem: Enabling Lifelong Experience Reuse for LLM Agents
Computation and Language
Summary
AI agents often need to learn from experience and remember skills to solve new problems. The authors created LifeMem, a method that helps these agents group past experiences by shared steps and recall useful skills when facing new tasks. This reduces forgetting old skills and helps the agents transfer knowledge across different environments. Their tests on many tasks showed that LifeMem lets AI better reuse what it learned over time.
What this means in practice
- •For robotics developers: Improve robots' ability to remember and reuse skills learned in past tasks across different environments.
- •For software engineers: Build adaptive software agents that reduce forgetting and better apply past experience to new challenges.
Authors
Yuli Qiu, Yutong Li, Wei Su, Zeming Liu, Wanxiang Che, Heyan Huang, Haifeng Wang, Yuang Guo
Abstract
Large language model agents are expected to continuously adapt to new tasks and environments over their lifetime by reusing past experience. However, existing memory-based agents struggle to transfer reusable experience across environments and suffer from catastrophic forgetting as experience accumulated. To address these challenges, we propose LifeMem, a lifelong learning framework that enables agents to transfer knowledge across multiple environments. During learning, LifeMem clusters accumulated interaction trajectories based on underlying workflows to extract reusable skills. When solving a new task at inference time, the agent recalls relevant skills and trajectories to guide actions. To validate our method, we conduct experiments across 10 environments and over 13k tasks with 2k newly annotated interaction trajectories. Results show that LifeMem enables effective experience reuse in lifelong learning, achieving both reduced forgetting on learned tasks and superior cross-task transfer. Further analysis reveals that task streaming impacts learning, while consolidating structurally similar trajectories within memory boosts performance.