Saved in:
Bibliographic Details
Main Authors: Kanjula, Karthik Reddy, Guthikonda, Surya, Alam, Nahid, Islam, Shayekh Bin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.06356
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915280069853184
author Kanjula, Karthik Reddy
Guthikonda, Surya
Alam, Nahid
Islam, Shayekh Bin
author_facet Kanjula, Karthik Reddy
Guthikonda, Surya
Alam, Nahid
Islam, Shayekh Bin
contents Pretraining datasets are foundational to the development of multimodal models, yet they often have inherent biases and toxic content from the web-scale corpora they are sourced from. In this paper, we investigate the prevalence of toxicity in LLaVA image-text pretraining dataset, examining how harmful content manifests in different modalities. We present a comprehensive analysis of common toxicity categories and propose targeted mitigation strategies, resulting in the creation of a refined toxicity-mitigated dataset. This dataset removes 7,531 of toxic image-text pairs in the LLaVA pre-training dataset. We offer guidelines for implementing robust toxicity detection pipelines. Our findings underscore the need to actively identify and filter toxic content - such as hate speech, explicit imagery, and targeted harassment - to build more responsible and equitable multimodal systems. The toxicity-mitigated dataset is open source and is available for further research.
format Preprint
id arxiv_https___arxiv_org_abs_2505_06356
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA
Kanjula, Karthik Reddy
Guthikonda, Surya
Alam, Nahid
Islam, Shayekh Bin
Computer Vision and Pattern Recognition
Pretraining datasets are foundational to the development of multimodal models, yet they often have inherent biases and toxic content from the web-scale corpora they are sourced from. In this paper, we investigate the prevalence of toxicity in LLaVA image-text pretraining dataset, examining how harmful content manifests in different modalities. We present a comprehensive analysis of common toxicity categories and propose targeted mitigation strategies, resulting in the creation of a refined toxicity-mitigated dataset. This dataset removes 7,531 of toxic image-text pairs in the LLaVA pre-training dataset. We offer guidelines for implementing robust toxicity detection pipelines. Our findings underscore the need to actively identify and filter toxic content - such as hate speech, explicit imagery, and targeted harassment - to build more responsible and equitable multimodal systems. The toxicity-mitigated dataset is open source and is available for further research.
title Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.06356