MemPro: Agentic Memory Systems as Evolvable Programs

📰 ArXiv cs.AI

arXiv:2606.00619v1 Announce Type: cross Abstract: Long-horizon autonomous agents require memory systems to retain historical information, track evolving states, and reuse relevant knowledge beyond finite context windows. Existing agentic memory systems typically follow a memory construction-retrieval (MCR) pipeline, but often adapt mainly the memory bank while keeping the surrounding pipeline fixed after deployment. This fixed-pipeline design struggles to handle heterogeneous task-specific failu

Published 2 Jun 2026
Read full paper → ← Back to Reads