Dataset Card for Adversarial GLUE

  Dataset Summary

Adversarial GLUE Benchmark (AdvGLUE) is a comprehensive robustness evaluation benchmark that focuses on the adversarial robustness evaluation of language models. It covers five natural language understanding tasks from the famous GLUE tasks and is an adversarial version of GLUE benchmark. AdvGLUE considers textual adversarial attacks from different perspectives and hierarchies, including word-level transformations… See the full description on the dataset page: https://huggingface.co/datasets/AI-Secure/adv_glue.

Adversarial Machine Learning TextFooler Dataset Overview This dataset, adversarial_machine_learning_textfooler_dataset.jsonl, is designed for research and evaluation in adversarial machine learning, specifically focusing on text-based adversarial attacks. It contains pairs of original and adversarial text samples, primarily generated using the TextFooler attack method, along with other adversarial techniques such as Homoglyph, Number Substitution, Typo, Emoji, Semantic Shift, Paraphrase, and… See the full description on the dataset page: https://huggingface.co/datasets/darkknight25/Adversarial_Machine_Learning_TextFooler_Dataset.

Adversarial Attacks Alignment Dataset

This dataset contains prompts and responses from various models, including accepted and rejected responses based on specific criteria. The dataset is designed to help in the study and development of adversarial attacks and alignment in reinforcement learning from human feedback (RLHF).

  Dataset Details

Prompts: Various prompts used to elicit responses from models. Accepted Responses: Responses that were accepted based on specific… See the full description on the dataset page: https://huggingface.co/datasets/yaswanth-iitkgp/Adversarial_Attacks_Alignment_Dataset.

Overview

This repo contains the text-to-image dataset of MMDT (Multimodal DecodingTrust). This research endeavor is designed to help researchers and practitioners better understand the capabilities, limitations, and potential risks involved in deploying the state-of-the-art Multimodal foundation models (MMFMs). This dataset focuses on the following six primary perspectives of trustworthiness, including safety, hallucination, fairness, privacy, adversarial robustness, and… See the full description on the dataset page: https://huggingface.co/datasets/AI-Secure/MMDecodingTrust-T2I.

Description: 👉 Download the dataset here The Abstract Paintings Dataset is designed to aid machine learning enthusiasts and researchers in experimenting with Generative Adversarial Networks (GANs) or other creative AI models, specifically tailored for generating abstract artworks. Initially inspired by the challenges encountered when using landscape images as training data, this dataset represents an alternative approach by focusing on abstract art. The dataset consists of images scraped, a… See the full description on the dataset page: https://huggingface.co/datasets/gtsaidata/Abstract-Paintings-Dataset.

Description: 👉 Download the dataset here Mathematical formula detection is a critical area in AI research, aiding in converting complex mathematical expressions from visual representations to machine-readable formats. This dataset aims to bridge the gap between handwritten or printed mathematical formulas and their digital counterparts, using deep learning models, particularly GANs (Generative Adversarial Networks). Download Dataset Dataset Overview The Math Formula Detection dataset includes… See the full description on the dataset page: https://huggingface.co/datasets/gtsaidata/MathFormulaDetectionDataset.