# Markedness in Visual Semantic AI

arXiv: https://arxiv.org/abs/2205.11378 · HTML: https://mlsanity.com/paper/2205.11378/

## Key facts

- Cited by 29 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 65 references that are in the map.
- First posted on arXiv 23 May 2022.

## Builds on (most cited first, 30 of 30)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks](https://mlsanity.com/paper/1908.02265/) (2019)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://mlsanity.com/paper/1908.03557/) (2019)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [Contrastive Representation Distillation](https://mlsanity.com/paper/1910.10699/) (2019)
- [SimVLM: Simple Visual Language Model Pretraining with Weak Supervision](https://mlsanity.com/paper/2108.10904/) (2021)
- [Contrastive Learning of Medical Visual Representations from Paired Images and Text](https://mlsanity.com/paper/2010.00747/) (2020)
- [The Woman Worked as a Babysitter: On Biases in Language Generation](https://mlsanity.com/paper/1909.01326/) (2019)
- [SLIP: Self-supervision meets Language-Image Pre-training](https://mlsanity.com/paper/2112.12750/) (2021)
- [On Measuring Social Biases in Sentence Encoders](https://mlsanity.com/paper/1903.10561/) (2019)
- [Multimodal datasets: misogyny, pornography, and malignant stereotypes](https://mlsanity.com/paper/2110.01963/) (2021)
- [Zero-Shot Learning Through Cross-Modal Transfer](https://mlsanity.com/paper/1301.3666/) (2013)
- [Detecting Emergent Intersectional Biases: Contextualized Word Embeddings Contain a Distribution of Human-like Biases](https://mlsanity.com/paper/2006.03955/) (2020)
- [Assessing Social and Intersectional Biases in Contextualized Word Representations](https://mlsanity.com/paper/1911.01485/) (2019)
- [Evaluating CLIP: Towards Characterization of Broader Capabilities and Downstream Implications](https://mlsanity.com/paper/2108.02818/) (2021)
- [Image Representations Learned With Unsupervised Pre-Training Contain Human-like Biases](https://mlsanity.com/paper/2010.15052/) (2020)
- [Are Gender-Neutral Queries Really Gender-Neutral? Mitigating Gender Bias in Image Search](https://mlsanity.com/paper/2109.05433/) (2021)
- [Predictive Inequity in Object Detection](https://mlsanity.com/paper/1902.11097/) (2019)
- [Low Frequency Names Exhibit Bias and Overfitting in Contextualizing Language Models](https://mlsanity.com/paper/2110.00672/) (2021)
- [Dodging Attack Using Carefully Crafted Natural Makeup](https://mlsanity.com/paper/2109.06467/) (2021)
- [Understanding the Representation and Representativeness of Age in AI Data Sets](https://mlsanity.com/paper/2103.09058/) (2021)
- [Contrastive Visual Semantic Pretraining Magnifies the Semantics of Natural Language Representations](https://mlsanity.com/paper/2203.07511/) (2022)

## Built on by (most cited first, 29 of 29)

- [Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale](https://mlsanity.com/paper/2211.03759/) (2022)
- [Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Models](https://mlsanity.com/paper/2305.18189/) (2023)
- [Fair Diffusion: Instructing Text-to-Image Generation Models on Fairness](https://mlsanity.com/paper/2302.10893/) (2023)
- [Inspecting the Geographical Representativeness of Images from Text-to-Image Models](https://mlsanity.com/paper/2305.11080/) (2023)
- [American == White in Multimodal Language-and-Image AI](https://mlsanity.com/paper/2207.00691/) (2022)
- [ChatGPT Perpetuates Gender Bias in Machine Translation and Ignores Non-Gendered Pronouns: Findings across Bengali and Five other Low-Resource Languages](https://mlsanity.com/paper/2305.10510/) (2023)
- [Survey of Social Bias in Vision-Language Models](https://mlsanity.com/paper/2309.14381/) (2023)
- [Gender Bias in Word Embeddings: A Comprehensive Analysis of Frequency, Syntax, and Semantics](https://mlsanity.com/paper/2206.03390/) (2022)
- [One Model to Edit Them All: Free-Form Text-Driven Image Manipulation with Semantic Modulations](https://mlsanity.com/paper/2210.07883/) (2022)
- [Would Deep Generative Models Amplify Bias in Future Models?](https://mlsanity.com/paper/2404.03242/) (2024)
- [Evaluating Biased Attitude Associations of Language Models in an Intersectional Context](https://mlsanity.com/paper/2307.03360/) (2023)
- [SANER: Annotation-free Societal Attribute Neutralizer for Debiasing CLIP](https://mlsanity.com/paper/2408.10202/) (2024)
- [BendVLM: Test-Time Debiasing of Vision-Language Embeddings](https://mlsanity.com/paper/2411.04420/) (2024)
- [Quantifying Bias in Text-to-Image Generative Models](https://mlsanity.com/paper/2312.13053/) (2023)
- [Social Perception of Faces in a Vision-Language Model](https://mlsanity.com/paper/2408.14435/) (2024)
- [Uncovering Bias in Face Generation Models](https://mlsanity.com/paper/2302.11562/) (2023)
- [DIG In: Evaluating Disparities in Image Generations with Indicators for Geographic Diversity](https://mlsanity.com/paper/2308.06198/) (2023)
- [Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models](https://mlsanity.com/paper/2510.03721/) (2025)
- [DiffusionWorldViewer: Exposing and Broadening the Worldview Reflected by Generative Text-to-Image Models](https://mlsanity.com/paper/2309.09944/) (2023)
- [From Cradle to Cane: A Two-Pass Framework for High-Fidelity Lifespan Face Aging](https://mlsanity.com/paper/2506.20977/) (2025)
- [Multi-Group Proportional Representation in Retrieval](https://mlsanity.com/paper/2407.08571/) (2024)
- [From Global to Local: Social Bias Transfer in CLIP](https://mlsanity.com/paper/2508.17750/) (2025)
- [Template-Based Probes Are Imperfect Lenses for Counterfactual Bias Evaluation in LLMs](https://mlsanity.com/paper/2404.03471/) (2024)
- [Neutral Prompts, Non-Neutral People: Quantifying Gender and Skin-Tone Bias in Gemini Flash 2.5 Image and GPT Image 1.5](https://mlsanity.com/paper/2602.12133/) (2026)
- [AnchorScore: A CLIP-Based Diagnostic of MLLM Annotation Difficulty](https://mlsanity.com/paper/2608.16690/) (2026)
- [Misgendering and Assuming Gender in Machine Translation when Working with Low-Resource Languages](https://mlsanity.com/paper/2401.13165/) (2024)
- ["I'm happy even though it's not real": GenAI Photo Editing as a Remembering Experience](https://mlsanity.com/paper/2602.03104/) (2026)
- [FAIR_XAI: Improving Multimodal Foundation Model Fairness via Explainability for Wellbeing Assessment](https://mlsanity.com/paper/2604.23786/) (2026)
- [Your Multimodal Speech Model Says I Have a Face for Radio](https://mlsanity.com/paper/2605.30472/) (2026)

## Abstract (arXiv)

We evaluate the state-of-the-art multimodal "visual semantic" model CLIP ("Contrastive Language Image Pretraining") for biases related to the marking of age, gender, and race or ethnicity. Given the option to label an image as "a photo of a person" or to select a label denoting race or ethnicity, CLIP chooses the "person" label 47.9% of the time for White individuals, compared with 5.0% or less for individuals who are Black, East Asian, Southeast Asian, Indian, or Latino or Hispanic. The model is more likely to rank the unmarked "person" label higher than labels denoting gender for Male individuals (26.7% of the time) vs. Female individuals (15.2% of the time). Age affects whether an individual is marked by the model: Female individuals under the age of 20 are more likely than Male individuals to be marked with a gender label, but less likely to be marked with an age label, while Female individuals over the age of 40 are more likely to be marked based on age than Male individuals. We also examine the self-similarity (mean pairwise cosine similarity) for each social group, where higher self-similarity denotes greater attention directed by CLIP to the shared characteristics (age, race, or gender) of the social group. As age increases, the self-similarity of representations of Female individuals increases at a higher rate than for Male individuals, with the disparity most pronounced at the "more than 70" age range. All ten of the most self-similar social groups are individuals under the age of 10 or over the age of 70, and six of the ten are Female individuals. Existing biases of self-similarity and markedness between Male and Female gender groups are further exacerbated when the groups compared are individuals who are White and Male and individuals who are Black and Female. Results indicate that CLIP reflects the biases of the language and society which produced its training data.