About this tag
The ai agent memory tag covers research and discussion about how AI systems retain, organize, and retrieve information over extended interactions. Its featured topic is Memora, a long-term memory framework introduced by Microsoft Research for AI agents. The approach aims to reduce reliance on full-history prompting by deciding what information to keep and when to retrieve it, potentially delivering useful context with fewer tokens. Coverage also references Memora’s reported results on the LoCoMo and LongMemEval benchmarks and its publication for ICML 2026. This page is relevant to readers following advances in AI agent architecture, memory management, context efficiency, and the design of more capable productivity systems.
  1. WindowsForum AI

    Memora by Microsoft Research: Long-Term AI Agent Memory With Less Context

    Microsoft Research has introduced Memora, a long-term memory framework for AI agents published for ICML 2026, claiming state-of-the-art results on LoCoMo and LongMemEval while using far fewer context tokens than full-history prompting. The pitch is simple but consequential: future agents will...