A Low-Latency Fraud Detection Layer for Detecting Adversarial Interaction Patterns in LLM-Powered Agents

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Yu, Sheldon, Sun, Yingcheng, Guo, Hanqing, McAuley, Julian, Tong, Qianqian
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917454176845824
author Yu, Sheldon
Sun, Yingcheng
Guo, Hanqing
McAuley, Julian
Tong, Qianqian
author_facet Yu, Sheldon
Sun, Yingcheng
Guo, Hanqing
McAuley, Julian
Tong, Qianqian
contents Large Language Model (LLM)-powered agents demonstrate strong capabilities in autonomous task execution, tool use, and multi-step reasoning. However, their increasing autonomy also introduces a new attack surface: adversarial interactions can manipulate agent behavior through direct prompt injection, indirect content attacks, and multi-turn escalation strategies. Existing defense strategies focus on prompt-level filtering and rule-based guardrails, which are often insufficient when risk emerges gradually across interaction sequences. In this work, we propose a complementary defense mechanism: a low-latency fraud detection layer for detecting adversarial interaction patterns in LLM-powered agents. Instead of determining whether a single prompt is malicious, our approach models risk over interaction trajectories using structured runtime features derived from prompt characteristics, session dynamics, tool usage, execution context, and fraud-inspired signals. The detection layer can be implemented using lightweight models leading to low-latency real-time deployments. To evaluate the framework, we construct a synthetic corpus of 12,000 multi-turn agent interactions generated from parameterized templates that simulate realistic agentic workflows. Using 42 structured features and an XGBoost classifier, our detector achieves over 9 times faster than LLM-based detectors. Through the experiment and ablation studies, our work suggests that interaction-level behavioral detection should become a core component of deployment-time defense for LLM-powered agents.
format Preprint
id arxiv_https___arxiv_org_abs_2605_01143
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Low-Latency Fraud Detection Layer for Detecting Adversarial Interaction Patterns in LLM-Powered Agents
Yu, Sheldon
Sun, Yingcheng
Guo, Hanqing
McAuley, Julian
Tong, Qianqian
Artificial Intelligence
Large Language Model (LLM)-powered agents demonstrate strong capabilities in autonomous task execution, tool use, and multi-step reasoning. However, their increasing autonomy also introduces a new attack surface: adversarial interactions can manipulate agent behavior through direct prompt injection, indirect content attacks, and multi-turn escalation strategies. Existing defense strategies focus on prompt-level filtering and rule-based guardrails, which are often insufficient when risk emerges gradually across interaction sequences. In this work, we propose a complementary defense mechanism: a low-latency fraud detection layer for detecting adversarial interaction patterns in LLM-powered agents. Instead of determining whether a single prompt is malicious, our approach models risk over interaction trajectories using structured runtime features derived from prompt characteristics, session dynamics, tool usage, execution context, and fraud-inspired signals. The detection layer can be implemented using lightweight models leading to low-latency real-time deployments. To evaluate the framework, we construct a synthetic corpus of 12,000 multi-turn agent interactions generated from parameterized templates that simulate realistic agentic workflows. Using 42 structured features and an XGBoost classifier, our detector achieves over 9 times faster than LLM-based detectors. Through the experiment and ablation studies, our work suggests that interaction-level behavioral detection should become a core component of deployment-time defense for LLM-powered agents.
title A Low-Latency Fraud Detection Layer for Detecting Adversarial Interaction Patterns in LLM-Powered Agents
topic Artificial Intelligence
url https://arxiv.org/abs/2605.01143