21 datasets found

The MMR benchmark is designed to evaluate MLLMs' comprehension of visual content and robustness against misleading questions, ensuring models truly leverage multimodal inputs...

The dataset used for testing the deep spectrum sensing framework with different signal types.

The dataset used for training and testing the deep spectrum sensing framework.

CIFAR-10-C and CIFAR-100-C are robustness benchmarks consisting of 19 corruptions types with five levels of severities.

Benchmarking robustness of deep learning classifiers using two-factor perturbation

Dataset bias is a significant problem in training fair classifiers. When attributes unrelated to classification exhibit strong biases towards certain classes, classifiers...

The MNIST dataset is a large dataset of handwritten digits, and the CIFAR-10 dataset is a dataset of images from 10 different classes.

Generative adversarial networks (GANs) are powerful gen-erative models, but usually suffer from instability and generalization problem which may lead to poor generations.

Multi-scale octave convolutional layers for robust speech recognition

Trajectory prediction plays an essential role in autonomous vehicles. While numerous strategies have been developed to enhance the robustness of trajectory prediction models,...

The LAV dataset is used to evaluate the robustness of the proposed Penalty-based Imitation Learning with Cross Semantics Generation approach.

The Imbalanced Gradients dataset is a benchmark for evaluating the robustness of deep neural networks.

The Squares dataset is a synthetic dataset composed of images with 224x224 pixels of centered squares with sides of 142 or 88 pixels.

The Spheres dataset is a synthetic dataset for which the data for each class lies in one of two concentric spheres, inspired by the work of Gilmer et al. (2018).

Human3.6M-C and HumanEva-I-C are constructed to motivate new robust 3D HPE solutions.

The proposed two-level group convolution is suitable for distributed memory computing and robust with respect to the large number of groups.

The dataset used in the paper is a multi-agent communication dataset, where agents learn to communicate with each other to achieve a common goal.

Robust exponential memory in Hopfield networks.

Associative content-addressable memory with exponentially many robust stable states and robust error correction.

The dataset used in this paper is MNIST, SVHN, CIFAR10, CIFAR100, and ImageNet.