Agent Memory Systems

Research systems such as Generative Agents use a memory stream, retrieval, reflection, and planning in a simulated-agent setting. Reflexion stores textual lessons from feedback for later attempts without changing model weights. These are research designs, not evidence that every memory system learns preferences correctly. Memory may be stale, irrelevant, or sensitive, so storage and retrieval need explicit policies. In the Generative Agents paper, each memory object contains a natural-language description, creation timestamp, and most-recent-access timestamp. Its retrieval score combines recency, relevance, and importance. These are properties of that paper’s implementation, not every agent-memory system. User-facing inspection, correction, and deletion are useful design controls for systems that retain personal facts. A useful classical distinction is semantic memory for general facts and learned knowledge versus episodic memory for personally experienced events situated in time and context. An agent memory design may represent both, but the labels describe different content and do not guarantee truth or permanence.