arXiv subjects

Robert Flood

Publications and source records attributed to Robert Flood.

3 recordsLinked to original sources

arXiv2026

CaMeLoT: CaMeL orchestrated with Temporal logic for static verification and liveness

LLM-based agents generate and execute multi-step plans that invoke external tools which can access private data or execute commands. In this setting, security is a property of the entire execution that a plan creates, not just any single step. The plan itself is a critical artefact that captures the tool calls, control flow, and data dependencies. We present CaMeLoT, a complement to CaMeL, an existing defence against prompt injection in tool-using LLM agents. CaMeLoT extends CaMeL by adding a static verification layer that checks an agent's plan before any tool is invoked. CaMeLoT translates a generated plan into a finite-state transition system, labels it with tool calls, provenance and taint information, and checks it against temporal policies expressed in CTL using the nuXmv model checker. Because verification happens before execution, unsafe plans are rejected without using LLM calls or tool calls, saving tokens that runtime could have cost, as well as the need to unwind changes or teardown temporary sandboxes. When a verification fails, the model checker returns a counterexample to give feedback to the agent to repair the plan. We evaluate CaMeLoT on policies derived from the AgentDojo benchmark, SOC workflows, and prompt-extraction experiments, showing that it verifies a broad class of temporal properties before execution while preserving CaMeL's runtime-checkable coverage.

cs.CR↗

arXiv2026

Detecting and Eliminating Neural Network Backdoors Through Active Paths with Application to Intrusion Detection

Machine learning backdoors have the property that the machine learning model should work as expected on normal inputs, but when the input contains a specific $\textit{trigger}$, it behaves as the attacker desires. Detecting such triggers has been proven to be extremely difficult. In this paper, we present a novel and explainable approach to detect and eliminate such backdoor triggers based on active paths found in neural networks. We present promising experimental evidence of our approach, which involves injecting backdoors into a machine learning model used for intrusion detection.

cs.CR↗

arXiv2020

Traffic Generation using Containerization for Machine Learning

The design and evaluation of data-driven network intrusion detection methods are currently held back by a lack of adequate data, both in terms of benign and attack traffic. Existing datasets are mostly gathered in isolated lab environments containing virtual machines, to both offer more control over the computer interactions and prevent any malicious code from escaping. This procedure however leads to datasets that lack four core properties: heterogeneity, ground truth traffic labels, large data size, and contemporary content. Here, we present a novel data generation framework based on Docker containers that addresses these problems systematically. For this, we arrange suitable containers into relevant traffic communication scenarios and subscenarios, which are subject to appropriate input randomization as well as WAN emulation. By relying on process isolation through containerization, we can match traffic events with individual processes, and achieve scalability and modularity of individual traffic scenarios. We perform two experiments to assess the reproducability and traffic properties of our framework, and demonstrate the usefulness of our framework on a traffic classification example.

cs.CR↗