Research map / Adversarial robustness and security / Black-box and transfer attacks
SADA: Semantic Adversarial Diagnostic Attacks for Autonomous Applications
Abdullah Hamdi, Matthias Mueller, Bernard Ghanem
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add boosting to BBGAN (ablation) | BBGAN on object detection · Object detection — 12-class average | AFR: 74.5% -> 80.5% | Improved |
| Add boosting to BBGAN (ablation) | BBGAN on object detection · Object detection — Bicycle | AFR: 44.0% -> 65.8% | Improved |
| Add boosting to BBGAN (ablation) | BBGAN on object detection · Object detection — Motorbike | AFR: 45.2% -> 82.0% | Improved |
| Add boosting to BBGAN (ablation) | BBGAN on object detection · Object detection — Truck | AFR: 90.8% -> 100% | Improved |
| Add boosting to BBGAN (ablation) | BBGAN on autonomous driving · CARLA — Straight | AFR: 100% -> 100% | No clear effect |
| Add boosting to BBGAN (ablation) | BBGAN on autonomous driving · CARLA — One Curve | AFR: 98.0% -> 100% | Improved |
| Add boosting to BBGAN (ablation) | BBGAN on autonomous driving · CARLA — Navigation | AFR: 98.0% -> 100% | Improved |
| Add boosting to BBGAN (ablation) | BBGAN on UAV racing · UAV racing — 3 anchors | AFR: 42.0% -> 86.0% | Improved |
| Add boosting to BBGAN (ablation) | BBGAN on UAV racing · UAV racing — 4 anchors | AFR: 94.0% -> 98.0% | Improved |
| Add boosting to BBGAN (ablation) | BBGAN on UAV racing · UAV racing — 5 anchors | AFR: 86.0% -> 92.0% | Improved |
| Use BBGAN (boost) instead of Bayesian optimization | YOLOv3 detector · Object detection — 12-class average | AFR: 56.1% -> 80.5% | Improved |
| Use BBGAN (boost) instead of GP Regression | UAV navigation agent · UAV racing — 5 anchors | AFR: 44.0% -> 92.0% | Improved |
Closest papers by meaning
- PhysGAN: Generating Physical-World-Resilient Adversarial Examples for Autonomous Driving (CVPR 2020 · similarity 0.87)
- Semi-supervised Semantics-guided Adversarial Training for Robust Trajectory Prediction (ICCV 2023 · similarity 0.87)
- Adversarial Backdoor Attack by Naturalistic Data Poisoning on Trajectory Prediction in Autonomous Driving (CVPR 2024 · similarity 0.85)
- Constructing Semantics-Aware Adversarial Examples with a Probabilistic Perspective (NeurIPS 2024 · similarity 0.85)
- AdvEDM: Fine-grained Adversarial Attack against VLM-based Embodied Agents (NeurIPS 2025 · similarity 0.84)
- BRUSLEATTACK: A QUERY-EFFICIENT SCORE- BASED BLACK-BOX SPARSE ADVERSARIAL ATTACK (ICLR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
