AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → memory today is an external module bolted onto a model. Metis compresses history into an in-model state read through memory attentionThe mechanism that lets a model weigh which earlier words matter for the word it's currently processing — the core operation of a transformer.Full definition → and updated by a single forward pass with frozen weights — a different cost and architecture profile than external stores.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
attention — The mechanism that lets a model weigh which earlier words matter for the word it's currently processing — the core operation of a transformer.
videoMemory and dreaming for self-learning agentsClaude
articleFilesystem-Based Memory for LLM Agents: Organization, Evolution, and SustainabilitySizhe Zhou, Sheldon Yu, Hui Wei, Junda Wu, Siru Ouyang, Yizhu Jiao, Shijia Pan, Julian McAuley, Yu Zhang, Tong Yu, Jiawei Han
videoAgent Memory EXPLAINED - Complete ArchitectureHugging Face