ContextBias: Controlled Evaluation of Bias Persistence Under Context Shift in Text-to-Image Models (EMNLP 2026) has been released.</p>\n","updatedAt":"2026-09-01T12:57:10.633Z","author":{"_id":"645b74b687c79b6ec0babd04","avatarUrl":"/avatars/21670f15de8999485f5ba3087dae14d1.svg","fullname":"kolli","name":"shaghayegh","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8570098280906677},"editors":["shaghayegh"],"editorAvatarUrls":["/avatars/21670f15de8999485f5ba3087dae14d1.svg"],"reactions":[],"isReport":false}},{"id":"6a977b90b9c98429e282cc61","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":379,"isUserFollowing":false},"createdAt":"2026-09-02T01:27:44.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"This is an automated message from the [Librarian Bot](https://huggingface.co/librarian-bots). I found the following papers similar to this paper. \n\nThe following papers were recommended by the Semantic Scholar API \n\n* [Investigating Social Bias in Narrative Image Generation](https://huggingface.co/papers/2608.01780) (2026)\n* [EmergencyBias: Bias in Text-to-Image Models under Emergency Scenarios](https://huggingface.co/papers/2608.00598) (2026)\n* [Debiasing Text-to-Image Evaluation via Implicit Cultural Alignment Reward Modeling](https://huggingface.co/papers/2607.15740) (2026)\n* [Locating and Controlling Implicit Personalization in Large Language Models](https://huggingface.co/papers/2608.11735) (2026)\n* [Beyond Surface Cues: Disentangling Sociocultural Signals in Multilingual LLMs](https://huggingface.co/papers/2608.23026) (2026)\n* [A Heuristic Perspective on Debiasing Language Models](https://huggingface.co/papers/2608.00622) (2026)\n* [Simile Understanding in Text-to-Image Models: An Evaluation Framework](https://huggingface.co/papers/2608.04750) (2026)\n\n\n Please give a thumbs up to this comment if you found it helpful!\n\n If you want recommendations for any Paper on Hugging Face checkout [this](https://huggingface.co/spaces/librarian-bots/recommend_similar_papers) Space\n\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: `@librarian-bot recommend`","html":"<p>This is an automated message from the <a href=\"https://huggingface.co/librarian-bots\">Librarian Bot</a>. I found the following papers similar to this paper. </p>\n<p>The following papers were recommended by the Semantic Scholar API </p>\n<ul>\n<li><a href=\"https://huggingface.co/papers/2608.01780\">Investigating Social Bias in Narrative Image Generation</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2608.00598\">EmergencyBias: Bias in Text-to-Image Models under Emergency Scenarios</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2607.15740\">Debiasing Text-to-Image Evaluation via Implicit Cultural Alignment Reward Modeling</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2608.11735\">Locating and Controlling Implicit Personalization in Large Language Models</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2608.23026\">Beyond Surface Cues: Disentangling Sociocultural Signals in Multilingual LLMs</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2608.00622\">A Heuristic Perspective on Debiasing Language Models</a> (2026)</li>\n<li><a href=\"https://huggingface.co/papers/2608.04750\">Simile Understanding in Text-to-Image Models: An Evaluation Framework</a> (2026)</li>\n</ul>\n<p> Please give a thumbs up to this comment if you found it helpful!</p>\n<p> If you want recommendations for any Paper on Hugging Face checkout <a href=\"https://huggingface.co/spaces/librarian-bots/recommend_similar_papers\">this</a> Space</p>\n<p> You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: <code>@librarian-bot recommend</code></p>\n","updatedAt":"2026-09-02T01:27:44.160Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":379,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7006304860115051},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.29847","authors":[{"_id":"6a96cac843140794820a7754","name":"Shaghayegh Kolli","hidden":false},{"_id":"6a96cac843140794820a7755","name":"Sina Emami","hidden":false},{"_id":"6a96cac843140794820a7756","name":"Moreno D'Incà","hidden":false},{"_id":"6a96cac843140794820a7757","name":"Pouyan Nejadi","hidden":false},{"_id":"6a96cac843140794820a7758","name":"Nicu Sebe","hidden":false},{"_id":"6a96cac843140794820a7759","name":"Massimiliano Mancini","hidden":false},{"_id":"6a96cac843140794820a775a","name":"Jana Diesner","hidden":false}],"publishedAt":"2026-08-30T00:00:00.000Z","submittedOnDailyAt":"2026-09-01T00:00:00.000Z","title":"ContextBias: Controlled Evaluation of Bias Persistence Under Context Shift in Text-to-Image Models","submittedOnDailyBy":{"_id":"645b74b687c79b6ec0babd04","avatarUrl":"/avatars/21670f15de8999485f5ba3087dae14d1.svg","isPro":false,"fullname":"kolli","user":"shaghayegh","type":"user","name":"shaghayegh"},"summary":"Text-to-image models learn associations between concepts - in the case of this paper, people's professions, which we refer to as roles - and visual attributes. These associations can underpin many observed forms of stereotypical bias. A key open question in this area is whether these associations are stable or change when visual representations of people in professional roles are placed in different prompted contexts. We introduce ContextBias, a controlled evaluation framework, and ContextBench, a benchmark spanning 92 roles and 1,656 semantically controlled prompts, designed to isolate the effect of contextual variation on role-linked visual representations. Evaluating four state-of-the-art models on 66,240 generated images, we find that placing a role in a semantically unrelated context does not suppress role-linked attributes; instead, cross-role attribute concentration increases (pooled BI +0.047). Demographic cues, characteristic garments, and role-specific tools remain highly prevalent across context-free, related, and unrelated conditions, and are robust to semantic prompt reformulation. Scene composition and camera framing show the greatest context-sensitivity. These findings reveal a form of stereotypical persistence that remains largely invisible to context-free evaluations, highlighting the need for controlled contextual variation in bias benchmarking. Code and dataset: https://huggingface.co/datasets/shaghayegh/ContextBias , https://github.com/Sina-Emami/ContextBias","upvotes":4,"discussionId":"6a96cac943140794820a775b","ai_summary":"Text-to-image models exhibit persistent role-linked visual stereotypes across varied contexts, with cross-role attribute concentration increasing rather than diminishing under unrelated prompts.","ai_keywords":["text-to-image models","ContextBias","ContextBench","stereotypical bias","cross-role attribute concentration","demographic cues","scene composition","camera framing"],"ai_summary_model":"thinkingmachines/Inkling-Small","organization":{"_id":"61fae781e68759322b9767be","name":"TUM","fullname":"Technical University of Munich","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1661167219960-629521a0f937190946e15d7f.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"645b74b687c79b6ec0babd04","avatarUrl":"/avatars/21670f15de8999485f5ba3087dae14d1.svg","isPro":false,"fullname":"kolli","user":"shaghayegh","type":"user"},{"_id":"63716b013d1bd47a4ec42e9a","avatarUrl":"/avatars/960a9fd329e018bfc533b5e6b245cc50.svg","isPro":false,"fullname":"Faeze","user":"Faeze","type":"user"},{"_id":"62cb464b6193ba3ced7e5082","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62cb464b6193ba3ced7e5082/CBovM4H7ESy3UuHk2etW9.jpeg","isPro":false,"fullname":"nikeghbal","user":"nafisehNik","type":"user"},{"_id":"63ac5701c21e60a3e9b58aa7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ac5701c21e60a3e9b58aa7/g6EX7diOpuA94R2ab-rZC.png","isPro":true,"fullname":"Dipankar Sarkar","user":"dipankarsarkar","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"61fae781e68759322b9767be","name":"TUM","fullname":"Technical University of Munich","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1661167219960-629521a0f937190946e15d7f.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.29847.md","query":{}}">
ContextBias: Controlled Evaluation of Bias Persistence Under Context Shift in Text-to-Image Models
Published on Aug 30
· Submitted by kolli on Sep 1 Abstract
Text-to-image models exhibit persistent role-linked visual stereotypes across varied contexts, with cross-role attribute concentration increasing rather than diminishing under unrelated prompts.
Text-to-image models learn associations between concepts - in the case of this paper, people's professions, which we refer to as roles - and visual attributes. These associations can underpin many observed forms of stereotypical bias. A key open question in this area is whether these associations are stable or change when visual representations of people in professional roles are placed in different prompted contexts. We introduce ContextBias, a controlled evaluation framework, and ContextBench, a benchmark spanning 92 roles and 1,656 semantically controlled prompts, designed to isolate the effect of contextual variation on role-linked visual representations. Evaluating four state-of-the-art models on 66,240 generated images, we find that placing a role in a semantically unrelated context does not suppress role-linked attributes; instead, cross-role attribute concentration increases (pooled BI +0.047). Demographic cues, characteristic garments, and role-specific tools remain highly prevalent across context-free, related, and unrelated conditions, and are robust to semantic prompt reformulation. Scene composition and camera framing show the greatest context-sensitivity. These findings reveal a form of stereotypical persistence that remains largely invisible to context-free evaluations, highlighting the need for controlled contextual variation in bias benchmarking. Code and dataset: https://huggingface.co/datasets/shaghayegh/ContextBias , https://github.com/Sina-Emami/ContextBias
Community
ContextBias: Controlled Evaluation of Bias Persistence Under Context Shift in Text-to-Image Models (EMNLP 2026) has been released.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2608.29847 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.29847 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2608.29847 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.