Images Speak Louder than Words: Understanding and Mitigating Bias in Vision-Language Model from a Causal Mediation Perspective

• URL: http://arxiv.org/abs/2407.02814v2
• Date: Mon, 07 Oct 2024 22:38:25 GMT
• Title: Images Speak Louder than Words: Understanding and Mitigating Bias in Vision-Language Model from a Causal Mediation Perspective
• Authors: Zhaotian Weng, Zijun Gao, Jerone Andrews, Jieyu Zhao,
• Abstract summary: Vision-language models pre-trained on extensive datasets can inadvertently learn biases by correlating gender information with objects or scenarios. We propose a framework that incorporates causal mediation analysis to measure and map the pathways of bias generation and propagation.
• Score: 13.486497323758226
• License:
• Abstract: Vision-language models (VLMs) pre-trained on extensive datasets can inadvertently learn biases by correlating gender information with specific objects or scenarios. Current methods, which focus on modifying inputs and monitoring changes in the model's output probability scores, often struggle to comprehensively understand bias from the perspective of model components. We propose a framework that incorporates causal mediation analysis to measure and map the pathways of bias generation and propagation within VLMs. This approach allows us to identify the direct effects of interventions on model bias and the indirect effects of interventions on bias mediated through different model components. Our results show that image features are the primary contributors to bias, with significantly higher impacts than text features, specifically accounting for 32.57% and 12.63% of the bias in the MSCOCO and PASCAL-SENTENCE datasets, respectively. Notably, the image encoder's contribution surpasses that of the text encoder and the deep fusion encoder. Further experimentation confirms that contributions from both language and vision modalities are aligned and non-conflicting. Consequently, focusing on blurring gender representations within the image encoder, which contributes most to the model bias, reduces bias efficiently by 22.03% and 9.04% in the MSCOCO and PASCAL-SENTENCE datasets, respectively, with minimal performance loss or increased computational demands.

Related papers

• Towards Deconfounded Image-Text Matching with Causal Inference [36.739004282369656]
  We propose an innovative Deconfounded Causal Inference Network (DCIN) for image-text matching task. DCIN decomposes the intra- and inter-modal confounders and incorporates them into the encoding stage of visual and textual features. It can learn causality instead of spurious correlations caused by dataset bias.
  arXiv  Detail & Related papers  (2024-08-22T11:04:28Z)
• Corpus Considerations for Annotator Modeling and Scaling [9.263562546969695]
  We show that the commonly used user token model consistently outperforms more complex models. Our findings shed light on the relationship between corpus statistics and annotator modeling performance.
  arXiv  Detail & Related papers  (2024-04-02T22:27:24Z)
• Debiasing Multimodal Large Language Models [61.6896704217147]
  Large Vision-Language Models (LVLMs) have become indispensable tools in computer vision and natural language processing. Our investigation reveals a noteworthy bias in the generated content, where the output is primarily influenced by the underlying Large Language Models (LLMs) prior to the input image. To rectify these biases and redirect the model's focus toward vision information, we introduce two simple, training-free strategies.
  arXiv  Detail & Related papers  (2024-03-08T12:35:07Z)
• 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)
• Debiasing Vision-Language Models via Biased Prompts [79.04467131711775]
  We propose a general approach for debiasing vision-language foundation models by projecting out biased directions in the text embedding. We show that debiasing only the text embedding with a calibrated projection matrix suffices to yield robust classifiers and fair generative models.
  arXiv  Detail & Related papers  (2023-01-31T20:09:33Z)
• Mitigating Representation Bias in Action Recognition: Algorithms and Benchmarks [76.35271072704384]
  Deep learning models perform poorly when applied to videos with rare scenes or objects. We tackle this problem from two different angles: algorithm and dataset. We show that the debiased representation can generalize better when transferred to other datasets and tasks.
  arXiv  Detail & Related papers  (2022-09-20T00:30:35Z)
• DASH: Visual Analytics for Debiasing Image Classification via User-Driven Synthetic Data Augmentation [27.780618650580923]
  Image classification models often learn to predict a class based on irrelevant co-occurrences between input features and an output class in training data. We call the unwanted correlations "data biases," and the visual features causing data biases "bias factors" It is challenging to identify and mitigate biases automatically without human intervention.
  arXiv  Detail & Related papers  (2022-09-14T00:44:41Z)
• General Greedy De-bias Learning [163.65789778416172]
  We propose a General Greedy De-bias learning framework (GGD), which greedily trains the biased models and the base model like gradient descent in functional space. GGD can learn a more robust base model under the settings of both task-specific biased models with prior knowledge and self-ensemble biased model without prior knowledge.
  arXiv  Detail & Related papers  (2021-12-20T14:47:32Z)
• Visual Recognition with Deep Learning from Biased Image Datasets [6.10183951877597]
  We show how biasing models can be applied to remedy problems in the context of visual recognition. Based on the (approximate) knowledge of the biasing mechanisms at work, our approach consists in reweighting the observations. We propose to use a low dimensional image representation, shared across the image databases.
  arXiv  Detail & Related papers  (2021-09-06T10:56:58Z)
• Understanding Adversarial Examples from the Mutual Influence of Images and Perturbations [83.60161052867534]
  We analyze adversarial examples by disentangling the clean images and adversarial perturbations, and analyze their influence on each other. Our results suggest a new perspective towards the relationship between images and universal perturbations. We are the first to achieve the challenging task of a targeted universal attack without utilizing original training data.
  arXiv  Detail & Related papers  (2020-07-13T05:00:09Z)

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.