Benchmarking the Spatial Robustness of DNNs via Natural and Adversarial Localized Corruptions

A new study has introduced an efficient refitting procedure for evaluating excess risk in empirical risk minimization under general convex loss functions. This method utilizes artificially modified pseudo-outcomes to create wild predictors, allowing for the computation of high-probability upper bounds without prior knowledge of the underlying function class complexity.

A recent study has proposed a novel approach for the unsupervised and source-free ranking of biomedical segmentation models, addressing the significant challenge of model selection in the biomedical community where data annotation is costly and time-consuming. This method builds on previous research linking model generalization and consistency under perturbation, aiming to facilitate the adoption of pre-trained models without the need for target dataset labels.

A recent study introduces a novel approach to visual emotion recognition (VER) by proposing the use of Partitioned Adaptive Contrastive Learning (PACL) to bridge the 'affective gap' between visual and textual modalities. This method leverages knowledge from pre-trained textual models to enhance the emotional perception capabilities of visual models, particularly in noisy social media contexts.

A new framework named FisheyeGaussianLift has been introduced, which enhances BEV (Bird's Eye View) semantic segmentation from fisheye camera imagery. This method addresses challenges such as non-linear distortion and occlusion by utilizing calibrated geometric unprojection and depth distribution estimation, achieving significant segmentation performance in complex environments.

A recent study has shown that semantic segmentation networks trained on specific lidar types struggle to generalize to new lidar systems without additional intervention. The research focuses on leveraging vision foundation models (VFMs) to enhance unsupervised domain adaptation for semantic segmentation of lidar point clouds, revealing key architectural insights for improving performance across different domains.