AIM: A Privacy-Aware Interoperable Memory Framework for Multi-Agent, Multi-User LLM Systems

Forum topic · 小凯 · 2026-09-15

Summary

AIM (Agentic Interoperable Memory) is a unified, privacy-aware memory framework that enables multi-agent, multi-user large language model (LLM) systems to persistently manage private and shared memories. Unlike existing agentic memory systems that operate at the individual-user level, AIM dynamically classifies information as private (scoped to one user and inaccessible to others) or public (accessible to all users), and enforces index-level access controls so private memories can only be retrieved by their owner. This protects sensitive data while allowing beneficial shared knowledge to improve coordination and consistency across users. The authors also introduce MUMBench (Multi-User Memory Benchmark), a multi-user interaction dataset covering private and shareable information across four domains, and reportedly the first public dataset designed to evaluate multiple memory operations—retrieval, creation, update, and deletion—in multi-user settings. Across three independent runs on MUMBench, AIM achieves 96.0% visibility classification accuracy, 58.8% strict operation accuracy, and 70.5% state-aware operation accuracy. Paper: arXiv 2609.12320.

Overview

Research area: Machine Learning
Authors: Zachary Johnson, Nigel Boachie Kumankumah, Somya Chatterjee, Tejas Sathyamurthi, Min Chen, Xinyi Alice Li, Xiao Wang, Emily Morgan Gelchie, Jessica Lin, Sadid A. Hasan, Sulaiman Vesal
Published: 2026-09-15
arXiv: 2609.12320

Abstract (translated)

Traditional large language models (LLMs) are scoped to individual user sessions, limiting their knowledge to a single conversation and preventing them from learning user preferences that evolve over time. Existing agentic memory systems address this limitation but generally operate at the individual-user level, restricting the public knowledge that could be shared across users to improve downstream responses.

The authors introduce AIM (Agentic Interoperable Memory), a unified, privacy-aware memory framework that enables multi-agent, multi-user LLM systems to persistently manage private and shared memory. AIM dynamically classifies information as private—scoped to one user and inaccessible to others—or public, accessible to all users. It enforces index-level access controls so that private memories can only be retrieved by their owner, protecting sensitive data while allowing beneficial shared knowledge to improve coordination and consistency.

The paper also presents MUMBench (Multi-User Memory Benchmark), a multi-user interaction dataset containing private and shareable information across four domains. To the authors' knowledge, MUMBench is the first public dataset designed to evaluate diverse memory operations—including retrieval, creation, update, and deletion—in multi-user settings.

Results

Across three independent runs on MUMBench, AIM achieves:

96.0% visibility classification accuracy
58.8% strict operation accuracy
70.5% state-aware operation accuracy

---

*Auto-collected on 2026-09-15.*