InfoResearchPeer-reviewed
MIST: Multi-Source Imperceptible Saliency-Guided Transferable Adversarial Attack Framework for LiDAR Point Cloud Perception in Remote Sensing
- Published
- Record updated
Summary
MIST is a transfer-based adversarial attack framework for LiDAR point cloud perception in remote sensing, published in Remote Sensing on 2026-11-15. It fuses source-model gradients, local geometric variation and predictive uncertainty into a multi-source saliency score that allocates perturbations, using reverse control for displacement limits and intensity scaling for update strength. Evaluated on ModelNet40, ScanObjectNN, KITTI and Waymo, it improves cross-model transferability while preserving geometric imperceptibility.
Topics
Related items
- MediumGemini AI agents repeatedly bypassed safeguards to attack real websiteSimilar attack · BleepingComputer
- InfoSTA: A Split Temporal Adversary Method for Evaluating Robustness of Multimodal RL-based Driving PoliciesSimilar attack · Arxiv (cs.RO + cs.CV security)
- InfoSmoothShift: Disrupting Visual Adversarial Attacks on LVLMs at Test TimeSimilar attack · Arxiv (cs.RO + cs.CV security)
- InfoAI cyber range: advancing education and training in adversarial and robust AISimilar attack · OpenAlex (peer-reviewed AI security)
- InfoCracks in Embeddings: Robustness and Stability of JEPA-Based Chest X-Ray RepresentationsSimilar attack · Arxiv (cs.RO + cs.CV security)