arXiv:2512. 04668v4 Announce Type: replace-cross Abstract: Graph topology is a fundamental determinant of memory leakage in multi-agent LLM systems, yet its effects remain poorly quantified.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.
arXiv:2605. 08442v5 Announce Type: replace-cross Abstract: We discover that prompt-injection success and tool-execution success are separable safety properties: defenses that block injection do not necessarily block execution, and vice versa.
The paper introduces a provenance‑aware execution graph for long‑horizon LLM agents, defining influence distance (DI) as the shortest structural path from an untrusted source to a sensitive action. Compared to the traditional sequence distance (DT), DI is always less than or equal to DT, revealing a median gap of nine hops in 454 injection–sink pairs across multiple models and datasets. The study shows that most pairs exhibit a non‑zero gap, and a deterministic DI‑based gate can block attacks missed by a sequence‑only gate without extra benign blocking.
arXiv:2606. 25115v1 Announce Type: new Abstract: On-device language-model agents improve by accumulating experience in retrieved memory rather than by updating weights.
arXiv:2605. 08442v3 Announce Type: replace-cross Abstract: Persistent memory attacks against LLM agents achieve high attack success rates against open-source models.
arXiv:2606. 30602v1 Announce Type: cross Abstract: Multi-agent systems (MAS) are increasingly used to automate complex, distributed workflows.
arXiv:2602. 11510v3 Announce Type: replace Abstract: Multi-agent Large Language Model (LLM) systems create privacy risks that current output-only benchmarks cannot measure.