arXiv:2606. 04656v1 Announce Type: cross Abstract: Object detection is a safety-critical component of autonomous driving.
Reliable motion classification is critical for autonomous driving, as false dynamic predictions of static objects can cascade into unnecessary planner interventions. Unstable bounding box predictions can lead to spurious velocity estimates in tracking and falsely predicted trajectories.
arXiv:2608. 09202v1 Announce Type: new Abstract: Modern autonomous vehicles are equipped with multiple sensors, such as cameras, LiDAR, and radar, for comprehensive environmental perception.
arXiv:2211.04340v2 Announce Type: replace Abstract: Autonomous driving systems must be capable of making quick decisions based on the perceived environment and specific driving conditions. Perception...
arXiv:2507. 08905v2 Announce Type: replace-cross Abstract: We explore the use of Hamiltonian Monte Carlo (HMC) sampling as a probabilistic last layer approach for deep neural networks (DNNs).
arXiv:2606. 26922v1 Announce Type: cross Abstract: Continuous driver monitoring in automated vehicles requires low-latency inference while avoiding unsafe decisions under uncertain driver states.
arXiv:2609.24668v1 Announce Type: new Abstract: Reliable uncertainty estimation is essential for deploying object detectors in autonomous systems operating in uncertain environments. Evidential Deep...
Reliable uncertainty estimation is essential for deploying object detectors in autonomous systems operating in uncertain environments. Evidential Deep Learning (EDL) provides a principled framework fo...
The paper introduces CLEAR, a lightweight, task‑agnostic post‑hoc method that enhances evidential robustness in deep learning models without retraining. CLEAR uses held‑out calibration data to map the geometry of the model’s latent space, then generates perturbation views at inference to detect latent conflict. When high conflict is found, CLEAR selectively reduces evidential strength while preserving evidence for latent‑consistent inputs, achieving significant improvements in OOD and adversarial AUROC on ImageNet→CUB and running much faster than competing methods.
The paper proposes a three-way open-set detection framework for autonomous navigation, classifying each detection as a known object, unknown object, or background based on a pretrained detector’s outputs. It introduces domain generalization and adaptation methods, evaluates them across various detector families and benchmarks, and demonstrates that this approach improves safety and efficiency in simulated navigation missions compared to binary detection methods.
arXiv:2606. 15479v1 Announce Type: cross Abstract: Steerable convolutional neural networks (Steerable-CNNs) guarantee SE(3)-equivariance by parameterizing kernels as linear combinations of steerable basis functions, but their deterministic nature precludes uncertainty quantification - limiting their use in settings where confidence estimates are essential.
The paper evaluates five vision‑language models on autonomous driving tasks under various visual input conditions, finding that visual corruption affects accuracy and confidence differently across models and datasets. It then tests Visual Evidence Augmentation (VEA) as an inference‑time technique to enhance reliability, observing mixed improvements depending on the model and setting.
arXiv:2606. 13818v1 Announce Type: new Abstract: This thesis investigates how Bayesian principles can deepen our understanding of modern deep learning systems.