Research

Paper

AI LLM March 04, 2026

Order Is Not Layout: Order-to-Space Bias in Image Generation

Authors

Yongkang Zhang, Zonglin Zhao, Yuechen Zhang, Fei Ding, Pei Li, Wenxuan Wang

Abstract

We study a systematic bias in modern image generation models: the mention order of entities in text spuriously determines spatial layout and entity--role binding. We term this phenomenon Order-to-Space Bias (OTS) and show that it arises in both text-to-image and image-to-image generation, often overriding grounded cues and causing incorrect layouts or swapped assignments. To quantify OTS, we introduce OTS-Bench, which isolates order effects with paired prompts differing only in entity order and evaluates models along two dimensions: homogenization and correctness. Experiments show that Order-to-Space Bias (OTS) is widespread in modern image generation models, and provide evidence that it is primarily data-driven and manifests during the early stages of layout formation. Motivated by this insight, we show that both targeted fine-tuning and early-stage intervention strategies can substantially reduce OTS, while preserving generation quality.

Metadata

arXiv ID: 2603.03714
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-03-04
Fetched: 2026-03-05 06:06

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.03714v1</id>\n    <title>Order Is Not Layout: Order-to-Space Bias in Image Generation</title>\n    <updated>2026-03-04T04:32:08Z</updated>\n    <link href='https://arxiv.org/abs/2603.03714v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.03714v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>We study a systematic bias in modern image generation models: the mention order of entities in text spuriously determines spatial layout and entity--role binding. We term this phenomenon Order-to-Space Bias (OTS) and show that it arises in both text-to-image and image-to-image generation, often overriding grounded cues and causing incorrect layouts or swapped assignments. To quantify OTS, we introduce OTS-Bench, which isolates order effects with paired prompts differing only in entity order and evaluates models along two dimensions: homogenization and correctness. Experiments show that Order-to-Space Bias (OTS) is widespread in modern image generation models, and provide evidence that it is primarily data-driven and manifests during the early stages of layout formation. Motivated by this insight, we show that both targeted fine-tuning and early-stage intervention strategies can substantially reduce OTS, while preserving generation quality.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.MM'/>\n    <published>2026-03-04T04:32:08Z</published>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Yongkang Zhang</name>\n    </author>\n    <author>\n      <name>Zonglin Zhao</name>\n    </author>\n    <author>\n      <name>Yuechen Zhang</name>\n    </author>\n    <author>\n      <name>Fei Ding</name>\n    </author>\n    <author>\n      <name>Pei Li</name>\n    </author>\n    <author>\n      <name>Wenxuan Wang</name>\n    </author>\n  </entry>"
}