pith. sign in

arxiv: 1605.06083 · v1 · pith:FWRCBIOEnew · submitted 2016-05-19 · 💻 cs.CL · cs.CV

Stereotyping and Bias in the Flickr30K Dataset

Emiel van Miltenburg This is my paper
classification 💻 cs.CL cs.CV
keywords datasetflickr30kassumptiondescriptionsaloneapplicationsbehindbias
0
0 comments X
read the original abstract

An untested assumption behind the crowdsourced descriptions of the images in the Flickr30K dataset (Young et al., 2014) is that they "focus only on the information that can be obtained from the image alone" (Hodosh et al., 2013, p. 859). This paper presents some evidence against this assumption, and provides a list of biases and unwarranted inferences that can be found in the Flickr30K dataset. Finally, it considers methods to find examples of these, and discusses how we should deal with stereotype-driven descriptions in future applications.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Adversarial Regularization for Visual Question Answering: Strengths, Shortcomings, and Side Effects

    cs.LG 2019-06 unverdicted novelty 5.0

    Adversarial regularization improves VQA performance on out-of-domain bias tests but introduces unstable gradients, reduced in-domain accuracy, and over-reliance on visual cues at the expense of linguistic information.