Mert D. Pesé
  • Home
  • Teaching
  • Publications
  • Lab
  • Research
  • Service
  • News
  • Publications
    • Meta-VLM: Which Metadata Do Vision-Language Models Need to Drive Safely?
    • User Perceptions of Deceptive In-Vehicle Infotainment User Interfaces
    • Relational Feature Distillation for Lightweight 3D Point Cloud Segmentation
    • On the Quantum Advantage in Black-Box Adversarial Attacks
    • On the Vulnerability of Lane Detectors to Physical Shadow
    • CANdy: Backward-Compatible CAN Message Authentication via Sub-Bit Injection on Commercial Controllers
    • GATE: Reliability-Gated Gaussian Evidence Fusion for Training-Free Test-Time Adaptation of Vision-Language Models
    • LLMCluster: Fuzzing Crash Triage using Generative AI
    • Evaluating the Robustness of Segmentation Models against Adversarial Patches in Off-Road Environments
    • Distilling Vision Language Model for Robust Traffic Sign Perception in Autonomous Vehicles
    • Short: CARdea: Two-Phase Plausibility-Based Vehicle-to-Vehicle Anomaly Detection System
    • Comparative Analysis of Patch Attack on VLM-Based Autonomous Driving Architectures
    • From MIRAGE to CLEAR - Component-Level Explainable Anomaly Reasoning for Autonomous Vehicle Perception Systems
    • NPNet: A Non-Parametric Network with Adaptive Gaussian-Fourier Positional Encoding for 3D Classification and Segmentation
    • Toward Inherently Robust VLMs Against Visual Perception Attacks
    • An Open-Source Testbed for Next-Generation In-Vehicle Zonal Architectures
    • Budget-Aware Adaptive Adversarial Patches for Black-Box Object Detection
    • Auditing Traffic-Sign Robustness via DDIM Inversion: Do Diffusion Latents Preserve Shadow Attacks?
    • FF3R: Feedforward Feature 3D Reconstruction from Unconstrained views
    • FlexMap: Generalized HD Map Construction from Flexible Camera Configurations
    • FuzzSense: Towards a Modular Fuzzing Framework for Autonomous Driving Software
    • Interpretable Adversarial Prompt Tuning via Semantic Concepts
    • Quantitative Evaluation of Git-Blockchain Synchronization Models for Trustworthy Software Provenance in Internet of Vehicles
    • SASA: Sequence-Aware Shadow Attacks via Attention Alignment for Traffic Sign Recognition
    • Secure Automotive Ethernet: Implementing and Benchmarking MACsec, IPsec, and TLS
    • SLNet: A Super-Lightweight Geometry-Adaptive Network for 3D Point Cloud Recognition
    • SoK: Security of the Image Processing Pipeline in Autonomous Vehicles
    • Understanding Adversarial Transferability in Vision-Language Models for Autonomous Driving - A Cross-Architecture Analysis
    • Enhancing Security Through Task Migration in Software-Defined Vehicles
    • Advancing Automotive Software Supply Chain Security: A Blockchain-Reproducible Build Approach
    • A First Look at Privacy Compliance of Zoom Apps
    • Attention-Aware Temporal Adversarial Shadows on Traffic Sign Sequences
    • Avoiding the Crash: A Vision-Language Model Evaluation of Critical Traffic Scenarios
    • Beyond the Glow: Understanding Luminescent Marker Behavior Against Autonomous Vehicle Perception Systems
    • David vs. Goliath: A comparative study of different-sized LLMs for code generation in the domain of automotive scenario generation
    • FedVLM: Scalable Personalized Vision-Language Models through Federated Learning
    • LLM-Powered Fuzz Testing of Automotive Diagnostic Protocols
    • MichiCAN: Spoofing and Denial-of-Service Protection using Integrated CAN Controllers
    • Security and data privacy of modern automobiles
    • Small Language Models on the Edge for Real-World Agentic Systems in Industry
    • Towards a Comprehensive Evaluation of Voltage-Based Fingerprinting for the CAN Bus
    • WIP: From Detection to Explanation: Using LLMs for Adversarial Scenario Analysis in Vehicles
    • An Initial Exploration of Employing Large Multimodal Models in Defending Against Autonomous Vehicles Attacks
    • Comparing Open-Source UDS Implementations Through Fuzz Testing
    • Fuzzing CAN vs. ROS: An Analysis of Single-Component vs. Dual-Component Fuzzing of Automotive Systems
    • AutoWatch: Learning Driver Behavior with Graphs for Auto Theft Detection and Situational Awareness
    • WIP: A First Look At Employing Large Multimodal Models Against Autonomous Vehicle Attacks
    • Analyzing Privacy Implications of Data Collection in Android Automotive OS
    • Contextualizing Security and Privacy of Software-Defined Vehicles: State of the Art and Industry Perspectives
    • Discovering New Shadow Patterns for Black-Box Attacks on Lane Detection of Autonomous Vehicles
    • Transforming In-Vehicle Network Intrusion Detection: VAE-based Knowledge Distillation Meets Explainable AI
    • An Overview of Security in Connected and Autonomous Vehicles
    • PRICAR: Privacy Framework for Vehicular Data Sharing with Third Parties
    • Using Phone Sensors to Augment Vehicle Reliability
    • A First Look at Android Automotive Privacy
    • Guess Which Car Type I Am Driving: Information Leak via Driving Apps
    • Achieving the safety and security of the end-to-end av pipeline
    • DETROIT: Data Collection, Translation and Sharing for Rapid Vehicular App Development
    • Bringing Practical Security to Vehicles
    • S2-CAN: Sufficiently Secure Controller Area Network
    • SPy: Car Steering Reveals Your Trip Route!
    • Security Analysis of Android Automotive
    • LibreCAN: Automated CAN Message Translator
    • Survey of Automotive Privacy Regulations and Privacy-Related Attacks
    • CarLab: Framework for Vehicular Data Collection and Processing
    • Context-aware Intrusion Detection in Automotive Control Systems
    • HW/SW Co-Design of an Automotive Embedded Firewall

Meta-VLM: Which Metadata Do Vision-Language Models Need to Drive Safely?

Sep 1, 2027·
David Fernandez
,
Pedram MohajerAnsari
,
Amir Salarpour
Mert D. Pesé
Mert D. Pesé
· 0 min read
Cite
Type
Conference paper
Publication
Winter Conference on Applications of Computer Vision (WACV'27)
Last updated on Sep 1, 2027
Mert D. Pesé
Authors
Mert D. Pesé
Assistant Professor
My research interests broadly lie in automotive security and privacy.
User Perceptions of Deceptive In-Vehicle Infotainment User Interfaces Jan 1, 2027 →

© 2026 Mert D. Pesé. This work is licensed under CC BY NC ND 4.0

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.