We present a novel task and dataset for evaluating the ability of vision and language models to conduct visio-linguistic compositional reasoning, which we call Winoground. Given two …
A Radford, JW Kim, C Hallacy… - International …, 2021 - proceedings.mlr.press
State-of-the-art computer vision systems are trained to predict a fixed set of predetermined object categories. This restricted form of supervision limits their generality and usability since …
AI models are increasingly applied in high-stakes domains like health and conservation. Data quality carries an elevated significance in high-stakes AI due to its heightened …
Data is a crucial component of machine learning. The field is reliant on data to train, validate, and test models. With increased technical capabilities, machine learning research has …
A Birhane, VU Prabhu - 2021 IEEE Winter Conference on …, 2021 - ieeexplore.ieee.org
In this paper we investigate problematic practices and consequences of large scale vision datasets (LSVDs). We examine broad issues such as the question of consent and justice as …
In response to growing concerns of bias, discrimination, and unfairness perpetuated by algorithmic systems, the datasets used to train and evaluate machine learning models have …
ID Raji, T Gebru, M Mitchell, J Buolamwini… - Proceedings of the …, 2020 - dl.acm.org
Although essential to revealing biased performance, well intentioned attempts at algorithmic auditing can have effects that may harm the very populations these measures are meant to …
Recent work identified a concerning trend of disproportional gender representation in research participants in Human–Computer Interaction (HCI). Motivated by the fact that …
E Simpson, B Semaan - Proceedings of the ACM on human-computer …, 2021 - dl.acm.org
Online communities provide spaces for people who are vulnerable and underserved to seek support and build community, such as LGBTQ+ people. Today, some online community …