#continuous learning

'A practical tutorial showing how to implement episodic and semantic memory for agentic systems so agents learn and improve across sessions.'

MemOS introduces a memory-centric operating system that transforms large language models by enabling structured, adaptive, and persistent memory management for continuous learning and better adaptability.

LifelongAgentBench introduces a novel benchmark for evaluating continuous learning in LLM-based agents, focusing on knowledge retention and adaptation across sequential tasks in dynamic environments.