34 datasets found

HONEST is a fairness dataset specifically designed to assess LMs' outputs' hurtfulness.

FAIRBELIEF is a language-agnostic analytical approach to capture and assess beliefs embedded in LMs.

DeBayes: a Bayesian method for debiasing network embeddings

The folktables dataset contains information about the mobility and income of individuals in California, with a sensitive feature of gender.

The dataset used in the paper is a set of UCI datasets, including adult, census, covertype, financial, joke, mushroom, and statlog.

The proposed framework was tested on four datasets, three of which are commonly used benchmarks. Two datasets (German credit and Adult income) come from the UCI ML repository...

The ACS-Income dataset is a dataset used for evaluating the fairness of machine learning models. It contains information about individuals and their income.

The dataset used in the paper for fair subgroup mixup for improving group fairness.

The dataset used in the paper for fair subgroup mixup for improving group fairness.

The COMPAS recidivism dataset contains 6,167 samples, and the task is to predict the recidivism of an individual based on criminal history, with the binary protected attribute...

The Bank Marketing dataset is a commonly used dataset in the fairness literature, containing information about individuals' demographic and economic characteristics.

The dataset used in this paper is the German dataset, which is a real-world UCI Machine Learning dataset extracted from a German bank for default prediction.

A commonly observed pattern in machine learning models is an underprediction of the target feature, with the model’s predicted target rate for members of a given category...

The dataset used in this paper is not explicitly described, but it is mentioned that it contains four DAGs representing well-known exposure-outcome relationships in the medical...