{"ID":2877043,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.20505","arxiv_id":"2508.20505","title":"Describe, Don't Dictate: Semantic Image Editing with Natural Language Intent","abstract":"Despite the progress in text-to-image generation, semantic image editing remains a challenge. Inversion-based algorithms unavoidably introduce reconstruction errors, while instruction-based models mainly suffer from limited dataset quality and scale. To address these problems, we propose a descriptive-prompt-based editing framework, named DescriptiveEdit. The core idea is to re-frame `instruction-based image editing' as `reference-image-based text-to-image generation', which preserves the generative power of well-trained Text-to-Image models without architectural modifications or inversion. Specifically, taking the reference image and a prompt as input, we introduce a Cross-Attentive UNet, which newly adds attention bridges to inject reference image features into the prompt-to-edit-image generation process. Owing to its text-to-image nature, DescriptiveEdit overcomes limitations in instruction dataset quality, integrates seamlessly with ControlNet, IP-Adapter, and other extensions, and is more scalable. Experiments on the Emu Edit benchmark show it improves editing accuracy and consistency.","short_abstract":"Despite the progress in text-to-image generation, semantic image editing remains a challenge. Inversion-based algorithms unavoidably introduce reconstruction errors, while instruction-based models mainly suffer from limited dataset quality and scale. To address these problems, we propose a descriptive-prompt-based edit...","url_abs":"https://arxiv.org/abs/2508.20505","url_pdf":"https://arxiv.org/pdf/2508.20505v1","authors":"[\"En Ci\",\"Shanyan Guan\",\"Yanhao Ge\",\"Yilin Zhang\",\"Wei Li\",\"Zhenyu Zhang\",\"Jian Yang\",\"Ying Tai\"]","published":"2025-08-28T07:45:08Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}