MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents
Long-lived AI agents require continuity across interactions, but continuity cannot be obtained by simply extending the prompt window. An agent must preserve useful prior experience, retrieve it selectively, distinguish personal context from external evidence, and revise memory when the underlying situation changes. We propose an architectural memory substrate organized along two orthogonal axes: a representational axis spanning structured records, vector representations, and graph relations; and a temporal axis spanning short-term traces, medium-term abstractions, and long-term semantic commit
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 60%sandst1/remind →
- PossiblePossibly related (embedding) · 58%ardhaecosystem/synapse →
- PossiblePossibly related (embedding) · 57%mem0ai/mem0 →
- PossiblePossibly related (embedding) · 57%basicmachines-co/basic-memory →
- PossiblePossibly related (embedding) · 56%We're building agents that can read millions of documents, but still forget a video they watched yesterday. →
- LinkedLinked via arxiv author · 85%Jizhizi Li →
“MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents”
- LinkedLinked via arxiv author · 85%Amy Shi-Nash →
“MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents”
- PossiblePossibly related (embedding) · 52%nossa-y/activity-frames →