中文
相关论文

相关论文: Joint Vision-Language Social Bias Removal for CLIP

200 篇论文

The Vision-Language Pre-training (VLP) models like CLIP have gained popularity in recent years. However, many works found that the social biases hidden in CLIP easily manifest in downstream tasks, especially in image retrieval, which can…

计算机视觉与模式识别 · 计算机科学 2024-05-31 Junyang Wang , Yi Zhang , Jitao Sang

Pre-trained vision-language models (VLMs), such as CLIP, have exhibited remarkable performance across various downstream tasks by aligning text and images in a unified embedding space. However, due to the imbalanced distribution of…

计算机视觉与模式识别 · 计算机科学 2024-10-15 Yunfan Yang , Chaoquan Jiang , Zhiyu Lin , Jinlin Xiao , Jiaming Zhang , Jitao Sang

Large vision-language contrastive models (VLCMs), such as CLIP, have become foundational, demonstrating remarkable success across a variety of downstream tasks. Despite their advantages, these models, akin to other foundational systems,…

计算机视觉与模式识别 · 计算机科学 2025-07-10 Haocheng Dai , Sarang Joshi

The rapid advancement of Vision-Language models (VLMs) has raised growing concerns that their black-box reasoning processes could lead to unintended forms of social bias. Current debiasing approaches focus on mitigating surface-level bias…

计算机视觉与模式识别 · 计算机科学 2026-03-02 Na Min An , Yoonna Jang , Yusuke Hirota , Ryo Hachiuma , Isabelle Augenstein , Hyunjung Shim

Vision-language models, like CLIP (Contrastive Language Image Pretraining), are becoming increasingly popular for a wide range of multimodal retrieval tasks. However, prior work has shown that large language and deep vision models can learn…

计算机视觉与模式识别 · 计算机科学 2024-11-05 Kimia Hamidieh , Haoran Zhang , Walter Gerych , Thomas Hartvigsen , Marzyeh Ghassemi

Vision-language model (VLM) embeddings have been shown to encode biases present in their training data, such as societal biases that prescribe negative characteristics to members of various racial and gender identities. VLMs are being…

计算机视觉与模式识别 · 计算机科学 2024-11-08 Walter Gerych , Haoran Zhang , Kimia Hamidieh , Eileen Pan , Maanas Sharma , Thomas Hartvigsen , Marzyeh Ghassemi

Contrastive Language-Image Pre-training (CLIP) has drawn increasing attention recently for its transferable visual representation learning. However, due to the semantic gap within datasets, CLIP's pre-trained image-text alignment becomes…

计算机视觉与模式识别 · 计算机科学 2023-08-11 Longtian Qiu , Renrui Zhang , Ziyu Guo , Ziyao Zeng , Zilu Guo , Yafeng Li , Guangnan Zhang

Multilingual vision-language models (VLMs) promise universal image-text retrieval, yet their social biases remain underexplored. We perform the first systematic audit of four public multilingual CLIP variants: M-CLIP, NLLB-CLIP,…

计算与语言 · 计算机科学 2025-11-20 Zahraa Al Sahili , Ioannis Patras , Matthew Purver

Vision language models such as CLIP have shown remarkable performance in zero shot classification, but remain susceptible to spurious correlations, where irrelevant visual features influence predictions. Existing debiasing methods often…

计算机视觉与模式识别 · 计算机科学 2025-11-04 Fangyu Wu , Yujun Cai

While Vision-Language Models (VLMs) have achieved remarkable performance across diverse downstream tasks, recent studies have shown that they can inherit social biases from the training data and further propagate them into downstream…

计算机视觉与模式识别 · 计算机科学 2026-03-16 Tangzheng Lian , Guanyu Hu , Yijing Ren , Dimitrios Kollias , Oya Celiktutan

Latent image representations arising from vision-language models have proved immensely useful for a variety of downstream tasks. However, their utility is limited by their entanglement with respect to different visual attributes. For…

计算机视觉与模式识别 · 计算机科学 2023-11-14 James Oldfield , Christos Tzelepis , Yannis Panagakis , Mihalis A. Nicolaou , Ioannis Patras

Vision-language models are growing in popularity and public visibility to generate, edit, and caption images at scale; but their outputs can perpetuate and amplify societal biases learned during pre-training on uncurated image-text pairs…

计算机视觉与模式识别 · 计算机科学 2023-05-25 Brandon Smith , Miguel Farinha , Siobhan Mackenzie Hall , Hannah Rose Kirk , Aleksandar Shtedritski , Max Bain

Recently, there have been breakthroughs in computer vision ("CV") models that are more generalizable with the advent of models such as CLIP and ALIGN. In this paper, we analyze CLIP and highlight some of the challenges such models pose.…

计算机视觉与模式识别 · 计算机科学 2021-08-09 Sandhini Agarwal , Gretchen Krueger , Jack Clark , Alec Radford , Jong Wook Kim , Miles Brundage

We explore the extent to which zero-shot vision-language models exhibit gender bias for different vision tasks. Vision models traditionally required task-specific labels for representing concepts, as well as finetuning; zero-shot models…

计算机视觉与模式识别 · 计算机科学 2023-01-27 Melissa Hall , Laura Gustafson , Aaron Adcock , Ishan Misra , Candace Ross

Recent advancements in Vision-Language Models (VLMs) have enabled complex multimodal tasks by processing text and image data simultaneously, significantly enhancing the field of artificial intelligence. However, these models often exhibit…

计算机视觉与模式识别 · 计算机科学 2024-10-30 Hoin Jung , Taeuk Jang , Xiaoqian Wang

Vision-language models (VLMs) have gained widespread adoption in both industry and academia. In this study, we propose a unified framework for systematically evaluating gender, race, and age biases in VLMs with respect to professions. Our…

计算机视觉与模式识别 · 计算机科学 2024-06-18 Ashutosh Sathe , Prachi Jain , Sunayana Sitaram

Many vision-language models (VLMs) that prove very effective at a range of multimodal task, build on CLIP-based vision encoders, which are known to have various limitations. We investigate the hypothesis that the strong language backbone in…

计算机视觉与模式识别 · 计算机科学 2025-09-22 Sho Takishita , Jay Gala , Abdelrahman Mohamed , Kentaro Inui , Yova Kementchedjhieva

Vision-language models can encode societal biases and stereotypes, but there are challenges to measuring and mitigating these multimodal harms due to lacking measurement robustness and feature degradation. To address these challenges, we…

Vision-Language Pre-training (VLP) models have achieved state-of-the-art performance in numerous cross-modal tasks. Since they are optimized to capture the statistical properties of intra- and inter-modality, there remains risk to learn…

计算机视觉与模式识别 · 计算机科学 2022-07-13 Yi Zhang , Junyang Wang , Jitao Sang

Machine learning models have been shown to inherit biases from their training datasets. This can be particularly problematic for vision-language foundation models trained on uncurated datasets scraped from the internet. The biases can be…

机器学习 · 计算机科学 2023-05-16 Ching-Yao Chuang , Varun Jampani , Yuanzhen Li , Antonio Torralba , Stefanie Jegelka
‹ 上一页 1 2 3 10 下一页 ›