← Pattern Library

Context & Memory

Memory

What agents remember across sessions, and how users see and edit it.

Observed Examples

Agents revise their opinions across repeated interaction rounds, implying persistent state that accumulates conviction over time.

arXiv · Physics of Agents: Statistical Mechanics Predicts Collective Behavior of AI Agents · Today

The study tests 7 memory conditions across 1,872 sessions to measure whether conversational LLMs naturally weave prior context into responses versus only recalling facts when directly queried.

arXiv · MemUse: Moving Memory Evaluation from Direct QA to Natural Integration in Long-Term Human-AI Conversation · Today

Matt discusses where memory fits into the future of agents as a distinct architectural concern in his Cloudflare Agents SDK work.

The Changelog · MCP on Code Mode (Interview) · Yesterday