Lifelong Learning of Large Language Model based Agents: A Roadmap
It provides a roadmap for researchers and practitioners to develop LLM agents that can adapt over time, mitigating catastrophic forgetting and improving long-term performance, which is incremental as it synthesizes existing ideas into a structured framework.
This survey addresses the lack of lifelong learning capabilities in large language model (LLM) agents by systematically summarizing techniques to enable continuous adaptation in dynamic environments, categorizing core components into perception, memory, and action modules.
Lifelong learning, also known as continual or incremental learning, is a crucial component for advancing Artificial General Intelligence (AGI) by enabling systems to continuously adapt in dynamic environments. While large language models (LLMs) have demonstrated impressive capabilities in natural language processing, existing LLM agents are typically designed for static systems and lack the ability to adapt over time in response to new challenges. This survey is the first to systematically summarize the potential techniques for incorporating lifelong learning into LLM-based agents. We categorize the core components of these agents into three modules: the perception module for multimodal input integration, the memory module for storing and retrieving evolving knowledge, and the action module for grounded interactions with the dynamic environment. We highlight how these pillars collectively enable continuous adaptation, mitigate catastrophic forgetting, and improve long-term performance. This survey provides a roadmap for researchers and practitioners working to develop lifelong learning capabilities in LLM agents, offering insights into emerging trends, evaluation metrics, and application scenarios. Relevant literature and resources are available at \href{this url}{https://github.com/qianlima-lab/awesome-lifelong-llm-agent}.