paper-with-me

홈 › Papers

Diagnosing and Correcting Concept Omission in Multimodal Diffusion Transformers

2026-05-14 · Kanghyun Baek, Jaihyun Lew, Chaehun Shin, Jungbeom Lee, Sungroh Yoon arxiv

Multimodal Diffusion Transformers (MM-DiTs) have achieved remarkable progress in text-to-image generation, yet they frequently suffer from concept omission, where specified objects or attributes fail to emerge in the generated image. By performing linear probing on text tokens, we demonstrate that text embeddings can distinguish a characteristic `omission signal' representing the absence of target concepts. Leveraging this insight, we propose Omission Signal Intervention (OSI), which amplifies the omission signal to actively catalyze the generation of missing concepts. Comprehensive experiments on FLUX.1-Dev and SD3.5-Medium demonstrate that OSI significantly alleviates concept omission even in extreme scenarios.

📄 PDF Abstract BibTeX arXiv:2605.14270

Code (0)

등록된 구현이 없습니다.

Tasks

Text-to-Image Generation

Similar Papers 제목 키워드 기반

Detecting and Correcting Learner Korean Particle Omission Errors

2013-10-01 · IJCNLP 2013 10 · Ross Israel, Markus Dickinson, Sun-Hee Lee
Grammatical Error DetectionMachine Translation

Delta-K: Boosting Multi-Instance Generation via Cross-Attention Augmentation

2026-03-10 · Zitong Wang, Zijun Shen, Haohao Xu, Zhengjie Luo 외 arxiv

While Diffusion Models excel in text-to-image synthesis, they often suffer from concept omission when synthesizing complex multi-instance scenes. Existing training-free methods attempt to resolve this by rescaling attent…

MMGPL: Multimodal Medical Data Analysis with Graph Prompt Learning

2023-12-22 · Liang Peng, Songyue Cai, Zongqian Wu, Huifang Shang 외

Prompt learning has demonstrated impressive efficacy in the fine-tuning of multimodal large models to a wide range of downstream tasks. Nonetheless, applying existing prompt learning methods for the diagnosis of neurolog…

Prompt LearningSemantic SimilaritySemantic Textual Similarity

Two Causes, Not One: Rethinking Omission and Fabrication Hallucinations in MLLMs

2025-08-30 · Guangzong Si, Hao Yin, Xianfei Li, Qing Ding 외 arxiv

Multimodal Large Language Models (MLLMs) have achieved impressive advances, yet object hallucination remains a persistent challenge. Existing methods, based on the flawed assumption that omission and fabrication hallucin…

What's Left Unsaid? Detecting and Correcting Misleading Omissions in Multimodal News Previews

2026-01-09 · Fanxiao Li, Jiaying Wu, Tingchao Fu, Dayang Li 외 arxiv

Even when factually correct, social-media news previews (image-headline pairs) can induce interpretation drift: by selectively omitting crucial context, they lead readers to form judgments that diverge from what the full…