collaborators

17 papers

cs.CR2026

CyberLLM: A Multi-Agent LLM Framework for Autonomous Detection and Guarded Response in Automotive Cybersecurity

Nenad Petrovic, Oussama Jeddou, Feres Ben Fraj +4

Software-Defined Vehicles (SDVs) expand the automotive attack surface across source code, runtime logs, and deployment topologies, while safety constraints forbid autonomous agents…

cs.MA2026

Scaling LLM-Driven Multi-Agent Systems: Design Principles and Architectural Scalability Analysis

Linus Sander, Fengjunjie Pan, Vahid Zolfaghari +3

The paper identifies four design principles for building scalable large‑language‑model‑driven multi‑agent systems, proposes a reference architecture based on a constrained directed…

cs.SE2026

LLM-Driven Approach to Modeling Tool Interoperability in Automotive Domain

Nenad Petrovic, Jiajie Zhang, Vahid Zolfaghari +1

The paper proposes using large language models to automate the mapping and merging of heterogeneous modeling tool metamodels in the automotive domain, enabling cross‑tool model int…

cs.LG2026

When LLMs Learn to Be Consistently Wrong: A Multi-Model Study of Linear Representations of Synthetic Deception

Vahideh Zolfaghari

Deceptive alignment, in which models maintain accurate internal representations while deliberately producing false outputs, remains a central challenge in AI safety. While strategi…

cs.SE2026

LLM-Empowered Functional Safety and Security by Design in Automotive Systems

Nenad Petrovic, Vahid Zolfaghari, Fengjunjie Pan +1

This paper presents LLM-empowered workflow to support Software Defined Vehicle (SDV) software development, covering the aspects of security-aware system topology design, as well as…

cs.CL2025

Cross-Platform Evaluation of Large Language Model Safety in Pediatric Consultations: Evolution of Adversarial Robustness and the Scale Paradox

Vahideh Zolfaghari

Background Large language models (LLMs) are increasingly deployed in medical consultations, yet their safety under realistic user pressures remains understudied. Prior assessments…