Quantifying Societal Bias Amplification in Image Captioning

• URL: http://arxiv.org/abs/2203.15395v1
• Date: Tue, 29 Mar 2022 09:42:11 GMT
• Title: Quantifying Societal Bias Amplification in Image Captioning
• Authors: Yusuke Hirota, Yuta Nakashima, Noa Garcia
• Abstract summary: We argue that, for image captioning, it is not enough to focus on the correct prediction of the protected attribute, and the whole context should be taken into account. We conduct extensive evaluation on traditional and state-of-the-art image captioning models, and surprisingly find that, by only focusing on the protected attribute prediction, bias mitigation models are unexpectedly amplifying bias.
• Score: 24.075869811508404
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We study societal bias amplification in image captioning. Image captioning models have been shown to perpetuate gender and racial biases, however, metrics to measure, quantify, and evaluate the societal bias in captions are not yet standardized. We provide a comprehensive study on the strengths and limitations of each metric, and propose LIC, a metric to study captioning bias amplification. We argue that, for image captioning, it is not enough to focus on the correct prediction of the protected attribute, and the whole context should be taken into account. We conduct extensive evaluation on traditional and state-of-the-art image captioning models, and surprisingly find that, by only focusing on the protected attribute prediction, bias mitigation models are unexpectedly amplifying bias.

Related papers

• Measuring directional bias amplification in image captions using predictability [13.041091740013808]
  We propose Directional Predictability Amplification in Captioning (DPAC) to measure bias amplification in ML datasets. DPAC measures directional bias amplification in captions, provides a better estimate of dataset bias, and is less sensitive to attacker models. Our experiments on the COCO captioning dataset show how DPAC is the most reliable metric to measure bias amplification in captions.
  arXiv  Detail & Related papers  (2025-03-10T21:50:58Z)
• GradBias: Unveiling Word Influence on Bias in Text-to-Image Generative Models [75.04426753720553]
  We propose a framework to identify, quantify, and explain biases in an open set setting. This pipeline leverages a Large Language Model (LLM) to propose biases starting from a set of captions. We show two variations of this framework: OpenBias and GradBias.
  arXiv  Detail & Related papers  (2024-08-29T16:51:07Z)
• Balancing the Picture: Debiasing Vision-Language Datasets with Synthetic Contrast Sets [52.77024349608834]
  Vision-language models can perpetuate and amplify societal biases learned during pre-training on uncurated image-text pairs from the internet. COCO Captions is the most commonly used dataset for evaluating bias between background context and the gender of people in-situ. We propose a novel dataset debiasing pipeline to augment the COCO dataset with synthetic, gender-balanced contrast sets.
  arXiv  Detail & Related papers  (2023-05-24T17:59:18Z)
• ImageCaptioner$^2$: Image Captioner for Image Captioning Bias Amplification Assessment [30.71835197717301]
  We introduce a new bias assessment metric, dubbed $ImageCaptioner2$, for image captioning. Instead of measuring the absolute bias in the model or the data, $ImageCaptioner2$ pay more attention to the bias introduced by the model w.r.t the data bias. In addition, we design a formulation for measuring the bias of generated captions as prompt-based image captioning.
  arXiv  Detail & Related papers  (2023-04-10T21:40:46Z)
• Measuring Representational Harms in Image Captioning [5.543867614999908]
  We present a set of techniques for measuring five types of representational harms, as well as the resulting measurements. Our goal was not to audit this image captioning system, but rather to develop normatively grounded measurement techniques. We discuss the assumptions underlying our measurement approach and point out when they do not hold.
  arXiv  Detail & Related papers  (2022-06-14T21:08:01Z)
• A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learning [55.96577490779591]
  Vision-language models can encode societal biases and stereotypes. There are challenges to measuring and mitigating these multimodal harms. We investigate bias measures and apply ranking metrics for image-text representations.
  arXiv  Detail & Related papers  (2022-03-22T17:59:04Z)
• Understanding and Evaluating Racial Biases in Image Captioning [18.184279793253634]
  We study bias propagation pathways within image captioning, focusing specifically on the COCO dataset. We demonstrate differences in caption performance, sentiment, and word choice between images of lighter versus darker-skinned people.
  arXiv  Detail & Related papers  (2021-06-16T01:07:24Z)
• Fine-Grained Image Captioning with Global-Local Discriminative Objective [80.73827423555655]
  We propose a novel global-local discriminative objective to facilitate generating fine-grained descriptive captions. We evaluate the proposed method on the widely used MS-COCO dataset.
  arXiv  Detail & Related papers  (2020-07-21T08:46:02Z)
• Mitigating Gender Bias Amplification in Distribution by Posterior Regularization [75.3529537096899]
  We investigate the gender bias amplification issue from the distribution perspective. We propose a bias mitigation approach based on posterior regularization. Our study sheds the light on understanding the bias amplification.
  arXiv  Detail & Related papers  (2020-05-13T11:07:10Z)
• Egoshots, an ego-vision life-logging dataset and semantic fidelity metric to evaluate diversity in image captioning models [63.11766263832545]
  We present a new image captioning dataset, Egoshots, consisting of 978 real life images with no captions. In order to evaluate the quality of the generated captions, we propose a new image captioning metric, object based Semantic Fidelity (SF)
  arXiv  Detail & Related papers  (2020-03-26T04:43:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.