Early Detection of Distributed Backdoors in Multi-Agent LLM Systems: A Characterization Study

arXiv:2607.24893 · cs.CR, cs.AI · Submitted 2026-07-27 · Read on arXiv

cs.CR, cs.AI

Submitted: 2026-07-27

Updated: 2026-10-02

Code: https://github.com/yibo-hu-lab/distributed-backdoor-early-warning

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Multi-Agent System — 9× in this paper · explained in 5 episodes
• Trace — 8× in this paper · explained in 2 episodes
• Accuracy — 7× in this paper · explained in 1 episode
• Benchmark — 4× in this paper · explained in 2 episodes
• Domain Shift — 4× in this paper · explained in 4 episodes
• Robustness — 4× in this paper · explained in 5 episodes
• Fine-Tuning — 3× in this paper · explained in 5 episodes
• Verifier — 3× in this paper · explained in 2 episodes
• Ablation — 2× in this paper · explained in 1 episode
• Benchmarking — 2× in this paper · explained in 3 episodes
• Generative AI — 2× in this paper · explained in 2 episodes
• AI Agents — 1× in this paper · explained in 2 episodes
• Chain-of-Thought Prompting — 1× in this paper · explained in 2 episodes
• Decode — 1× in this paper · explained in 2 episodes
• In International Conference on Learning Representations — 1× in this paper
• Meeting of the Association for Computational Linguistics — 1× in this paper
• Perplexity — 1× in this paper · explained in 3 episodes
• Quantization — 1× in this paper · explained in 12 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Reward Shaping — 1× in this paper · explained in 2 episodes
• Unified Framework — 1× in this paper · explained in 5 episodes

Sources

• AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security
• TrojanRAG: Retrieval-Augmented Generation Can Be Backdoor Driver in Large Language Models
• BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents
• SkillTrojan: Backdoor Attacks on Skill-Based Agent Systems
• PrefixGuard: Online Failure Warning and Trace-Grounded Diagnosis for LLM Agents
• HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark
• AgentForesight: Online Auditing for Early Failure Prediction in Multi-Agent Systems
• Collaborative Shadows: Distributed Backdoor Attacks in LLM-Based Multi-Agent Systems
• DemonAgent: Dynamically Encrypted Multi-Backdoor Implantation Attack on LLM-based Agent

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs