{"ID":2832930,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.04668","arxiv_id":"2512.04668","title":"Topology Matters: Measuring Memory Leakage in Multi-Agent LLMs","abstract":"Graph topology is a fundamental determinant of memory leakage in multi-agent LLM systems, yet its effects remain poorly quantified. We introduce MAMA (Multi-Agent Memory Attack), a framework that measures how network structure shapes leakage. MAMA operates on synthetic documents containing labeled Personally Identifiable Information (PII) entities, from which we generate sanitized task instructions. We execute a two-phase protocol: Engram (seeding private information into a target agent's memory) and Resonance (multi-round interaction where an attacker attempts extraction). Over 10 rounds, we measure leakage as exact-match recovery of ground-truth PII from attacker outputs. We evaluate six canonical topologies (complete, ring, chain, tree, star, star-ring) across $n\\in\\{4,5,6\\}$, attacker-target placements, and base models. Results are consistent: denser connectivity, shorter attacker-target distance, and higher target centrality increase leakage; most leakage occurs in early rounds and then plateaus; model choice shifts absolute rates but preserves topology ordering; spatiotemporal/location attributes leak more readily than identity credentials or regulated identifiers. We distill practical guidance for system design: favor sparse or hierarchical connectivity, maximize attacker-target separation, and restrict hub/shortcut pathways via topology-aware access control.","short_abstract":"Graph topology is a fundamental determinant of memory leakage in multi-agent LLM systems, yet its effects remain poorly quantified. We introduce MAMA (Multi-Agent Memory Attack), a framework that measures how network structure shapes leakage. MAMA operates on synthetic documents containing labeled Personally Identifiab...","url_abs":"https://arxiv.org/abs/2512.04668","url_pdf":"https://arxiv.org/pdf/2512.04668v3","authors":"[\"Jinbo Liu\",\"Defu Cao\",\"Yifei Wei\",\"Tianyao Su\",\"Yuan Liang\",\"Yushun Dong\",\"Yan Liu\",\"Yue Zhao\",\"Xiyang Hu\"]","published":"2025-12-04T11:00:49Z","proceeding":"cs.CR","tasks":"[\"cs.CR\",\"cs.AI\",\"cs.CL\"]","methods":"[\"Large Language Model\"]","has_code":false}