Five-Tier Memory Cascade
Stage agent memory across five tiers — sensory, working, short-term, episodic, long-term — with explicit promotion and decay between each, matching each tier to its natural timescale.
Intent & Description
🎯 Intent
Stage agent memory across sensory, working, short-term, episodic, and long-term tiers with explicit promotion and decay between them.
📋 Context
A long-running agent accumulates information at very different timescales. Some observations are one-tick-only (“the user just clicked save”); some are day-level patterns; some are month-level preferences; some are stable identity facts. A flat single-tier store cannot represent these differences in age, decay rate, or relevance horizon.
💡 Solution
Five tiers — Sensory: raw input per tick. Working: top-N items in active focus (Global Workspace Theory, ≤7 items). Short-term: recent verbatim (1–7 days). Episodic: compressed summaries (5–10×). Long-term: distilled rules and insights. Compaction promotes upward on a schedule; decay archives downward; rehearsal lifts archived items back when re-attended.
Real-world Use Case
- A flat append-only log is collapsing signal across timescales (sensory, working, recent, episodic, distilled).
- Promotion and decay between tiers can be implemented on a schedule.
- Working memory needs an explicit cap (e.g. ≤7 items per Global Workspace Theory).
Source
📌 TL;DR
Model memory as five tiers with explicit promotion and decay — each timescale gets its own storage strategy instead of everything collapsing into one flat append-only log.
Advantages
- Each tier is optimized for its natural timescale — sensory is cheap and ephemeral, long-term is durable and distilled.
- The memory hierarchy is inspectable and maps to well-understood cognitive science vocabulary.
Disadvantages
- Architecturally heavy — only earns its seat in genuinely long-running agents; overkill for short sessions.
- Tuning promotion thresholds and decay rates between tiers is empirical work per deployment.