2026-07-23 04:00 UTCOriginal source2 min readUpdated: 2026-07-23 08:18 UTC

OpenEvoShield: Dual Non-Stationary Continual Defense for Open-World Multi-Agent System Attacks

OpenEvoShield is a continual defense framework for LLM-based multi-agent systems that addresses dual dynamics of attack adaptation and normal behavior drift, using an asymmetric rate controller, dynamic boundary updater, EWC-regularized policy ensemble, and energy-based detector to detect unknown attacks with low false positives across 100 deployment rounds.

SourcearXiv AIAuthor: Litian Zhang, Chaozhuo Li, Yuting Zhang, Zejian Chen, Bingyu Yan, Qiwei Ye

Article intelligence

EngineersIntermediate

Key points

LLM multi-agent systems face dual dynamics: adversaries refine attack strategies and normal behavior drifts; existing defenses assume a closed world and degrade quickly.
OpenEvoShield features three modules: asymmetric rate controller decouples fast and slow learning, normal-boundary updater maintains dynamic boundaries, and EWC-regularized policy ensemble enables fast adaptation.
An energy-based detector fuses node, subgraph, and graph evidence to classify novel attacks as out-of-distribution.
Experiments over five benchmarks and four topologies show OpenEvoShield outperforms static and continual baselines.

Why it matters

This matters because LLM multi-agent systems face dual dynamics: adversaries refine attack strategies and normal behavior drifts; existing defenses assume a closed world and degrade quickly.

Technical impact

May affect model selection, inference cost, product capability, and evaluation benchmarks.

This panel is AI-generated and reviewed for accuracy.

-->

[Submitted on 13 May 2026]

Title:OpenEvoShield: Dual Non-Stationary Continual Defense for Open-World Multi-Agent System Attacks

View a PDF of the paper titled OpenEvoShield: Dual Non-Stationary Continual Defense for Open-World Multi-Agent System Attacks, by Litian Zhang and 5 other authors

View PDF HTML (experimental)

Abstract:LLM-based multi-agent systems (LLM-MAS) are increasingly deployed in safety-critical applications, where adversaries inject malicious instructions through inter-agent communication to propagate harmful behaviors. Unlike static threats, these attacks are doubly dynamic: adversaries refine injection strategies against deployed defenses while normal-agent behavior drifts with system expansion. Existing defenses treat deployment as a closed-world problem and degrade rapidly once either distribution shifts beyond training coverage. We propose OpenEvoShield, a co-evolutionary continual defense framework for LLM-MAS. An asymmetric rate controller (M1) decouples fast attack-side and slow normal-side learning rates from dual drift signals. A normal-boundary updater (M2) maintains a dynamic behavioral boundary at the slow rate, while an EWC-regularized policy ensemble (M3) fast-adapts without catastrophic forgetting. An energy-based multi-granularity detector (M4) fuses node-, subgraph-, and graph-level evidence to classify novel attacks as out-of-distribution. Experiments over 100 deployment rounds across five benchmarks and four MAS topologies show that OpenEvoShield outperforms static and continual baselines, detecting most previously unseen attacks while keeping false positive rates low.

Comments: 29 pages, 5 figures, 14 tables

Subjects:

Artificial Intelligence (cs.AI)

Cite as: arXiv:2607.19351 [cs.AI]

(or arXiv:2607.19351v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2607.19351

arXiv-issued DOI via DataCite

Submission history

From: Zejian Chen [view email] [v1] Wed, 13 May 2026 04:28:46 UTC (1,573 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled OpenEvoShield: Dual Non-Stationary Continual Defense for Open-World Multi-Agent System Attacks, by Litian Zhang and 5 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.AI

new | recent | 2026-07

Change to browse by:

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)