paper-with-me

홈 › Papers

Task Bias in Vision-Language Models

2022-12-08 · Sachit Menon, Ishaan Preetam Chandratreya, Carl Vondrick

Incidental supervision from language has become a popular approach for learning generic visual representations that can be prompted to perform many recognition tasks in computer vision. We conduct an in-depth exploration of the CLIP model and show that its visual representation is often strongly biased towards solving some tasks more than others. Moreover, which task the representation will be biased towards is unpredictable, with little consistency across images. To resolve this task bias, we show how to learn a visual prompt that guides the representation towards features relevant to their task of interest. Our results show that these visual prompts can be independent of the input image and still effectively provide a conditioning mechanism to steer visual representations towards the desired task.

📄 PDF Abstract BibTeX arXiv:2212.04412

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

CLIP Contrastive Language-Image Pre-training (CLIP), consisting of a simplified version of ConVIRT trained from scratch, is an efficient method of image representation learning…

Similar Papers 제목 키워드 기반

Think Before You Act: A Two-Stage Framework for Mitigating Gender Bias Towards Vision-Language Tasks

2024-05-27 · Yunqi Zhang, Songda Li, Chunyuan Deng, Luyi Wang 외

Gender bias in vision-language models (VLMs) can reinforce harmful stereotypes and discrimination. In this paper, we focus on mitigating gender bias towards vision-language tasks. We identify object hallucination as the …

HallucinationObject Hallucination

Vision-Language Models Performing Zero-Shot Tasks Exhibit Gender-based Disparities

2023-01-26 · Melissa Hall, Laura Gustafson, Aaron Adcock, Ishan Misra 외

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; …

image-classificationImage Classificationobject-detectionObject Detection+3

VisBias: Measuring Explicit and Implicit Social Biases in Vision Language Models

2025-03-10 · Jen-tse Huang, Jiantong Qin, Jianping Zhang, Youliang Yuan 외

This research investigates both explicit and implicit social biases exhibited by Vision-Language Models (VLMs). The key distinction between these bias types lies in the level of awareness: explicit bias refers to conscio…

Image DescriptionMultiple-choice

Med-UniC: Unifying Cross-Lingual Medical Vision-Language Pre-Training by Diminishing Bias

2023-05-31 · NeurIPS 2023 11 · Zhongwei Wan, Che Liu, Mi Zhang, Jie Fu 외

The scarcity of data presents a critical obstacle to the efficacy of medical visionlanguage pre-training (VLP). A potential solution lies in the combination of datasets from various language communities. Nevertheless, th…

Disentanglement

Freeze and Reveal: Exposing Modality Bias in Vision-Language Models

2025-08-10 · Vivek Hruday Kavuri, Vysishtya Karanam, Venkata Jahnavi Venkamsetty, Kriti Madumadukala 외 arxiv

Vision Language Models achieve impressive multi-modal performance but often inherit gender biases from their training data. This bias might be coming from both the vision and text modalities. In this work, we dissect the…

Data Augmentation