{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/ultraedit-instruction-based-fine-grained","title":"UltraEdit: Instruction-based Fine-Grained Image Editing at Scale","arxiv_id":"2407.05282","date":"2024-07-07","proceeding":null,"authors":["Haozhe Zhao","Xiaojian Ma","Liang Chen","Shuzheng Si","Rujie Wu","Kaikai An","Peiyu Yu","Minjia Zhang","Qing Li","Baobao Chang"],"abstract":"This paper presents UltraEdit, a large-scale (approximately 4 million editing samples), automatically generated dataset for instruction-based image editing. Our key idea is to address the drawbacks in existing image editing datasets like InstructPix2Pix and MagicBrush, and provide a systematic approach to producing massive and high-quality image editing samples. UltraEdit offers several distinct advantages: 1) It features a broader range of editing instructions by leveraging the creativity of large language models (LLMs) alongside in-context editing examples from human raters; 2) Its data sources are based on real images, including photographs and artworks, which provide greater diversity and reduced bias compared to datasets solely generated by text-to-image models; 3) It also supports region-based editing, enhanced by high-quality, automatically produced region annotations. Our experiments show that canonical diffusion-based editing baselines trained on UltraEdit set new records on MagicBrush and Emu-Edit benchmarks. Our analysis further confirms the crucial role of real image anchors and region-based editing data. The dataset, code, and models can be found in https://ultra-editing.github.io.","url_abs":"https://arxiv.org/abs/2407.05282v2","url_pdf":"https://arxiv.org/pdf/2407.05282v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"ultraedit-instruction-based-fine-grained","repo_url":"https://github.com/pkunlp-icler/ultraedit","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"jax","reach":{"status":"ok"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"image-editing","task_name":"Image Editing"}],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-editing-on-imgedit-data","task":"Image Editing","dataset":"ImgEdit-Data","model":"UltraEdit","rank_in_archive_order":6,"of":9,"metrics":{"Action":"2.98","Add":"3.44","Adjust":"2.81","Background":"2.83","Extract":"2.13","Hybrid":"1.91","Overall":"2.70","Remove":"1.45","Replace":"2.96","Style":"3.76"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2407.05282","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.05282"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pkunlp-icler/ultraedit","reach":{"status":"ok"}}],"summary":{"ran":9,"ran_draft_wrong":1,"unverified":7},"by_repo_kind":{"official":{"samples":17,"ran":10,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":17,"samples":[{"code_sha256_prefix":"a56d6564fd11974d","entry":"check_mask_size","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/data_generation.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/data_generation.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a56d6564fd11974d"}},{"code_sha256_prefix":"629773a88a55f9e6","entry":"convert_to_np","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"traning/train_sdxl_pix2pix.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/traning/train_sdxl_pix2pix.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"629773a88a55f9e6"}},{"code_sha256_prefix":"1c34ef7c86b190fb","entry":"download_image","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"traning/train_sdxl_pix2pix.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/traning/train_sdxl_pix2pix.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1c34ef7c86b190fb"}},{"code_sha256_prefix":"56b8d77956f342de","entry":"find_contours_number","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/data_generation.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/data_generation.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"56b8d77956f342de"}},{"code_sha256_prefix":"c00e5e79dd083932","entry":"generate_Contours_mask","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/data_generation.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/data_generation.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c00e5e79dd083932"}},{"code_sha256_prefix":"0a26bb2a52f65689","entry":"get_time_words_attention_alpha","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/processors.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/processors.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0a26bb2a52f65689"}},{"code_sha256_prefix":"dd89dacf389e9e01","entry":"import_model_class_from_model_name_or_path","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"traning/train_sdxl_pix2pix.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/traning/train_sdxl_pix2pix.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dd89dacf389e9e01"}},{"code_sha256_prefix":"bea2d776a332f2b0","entry":"rescale_noise_cfg","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/prompt_to_prompt_pipeline.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/prompt_to_prompt_pipeline.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"bea2d776a332f2b0"}},{"code_sha256_prefix":"dce9df5004a68cce","entry":"tokenize_prompt","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"traning/train_sd3_pix2pix.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/traning/train_sd3_pix2pix.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dce9df5004a68cce"}},{"code_sha256_prefix":"dd498a88e593fe37","entry":"update_alpha_time_word","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/processors.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/processors.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dd498a88e593fe37"}},{"code_sha256_prefix":"be76f458330f9eec","entry":"check_tags_chinese","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/Grounded-Segment-Anything/automatic_label_ram_demo.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/Grounded-Segment-Anything/automatic_label_ram_demo.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"be76f458330f9eec"}},{"code_sha256_prefix":"93889c3f8ba8ba23","entry":"create_controller","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/processors.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/processors.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"93889c3f8ba8ba23"}},{"code_sha256_prefix":"a91b81d1e36f84e1","entry":"create_controller","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/util.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a91b81d1e36f84e1"}},{"code_sha256_prefix":"1b3f076c87d15223","entry":"encode_prompt","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"traning/train_sd3_pix2pix.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/traning/train_sd3_pix2pix.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1b3f076c87d15223"}},{"code_sha256_prefix":"d893c774b1f34088","entry":"generate_caption","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/Grounded-Segment-Anything/automatic_label_demo.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/Grounded-Segment-Anything/automatic_label_demo.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d893c774b1f34088"}},{"code_sha256_prefix":"c0648fc5a77afd71","entry":"generate_tags","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"data_generation/Grounded-Segment-Anything/automatic_label_demo.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/data_generation/Grounded-Segment-Anything/automatic_label_demo.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c0648fc5a77afd71"}},{"code_sha256_prefix":"74ab2d24d15218ee","entry":"load_text_encoders","repo":"pkunlp-icler/ultraedit","repo_kind":"official","path":"traning/train_sd3_pix2pix.py","file_url":"https://github.com/pkunlp-icler/ultraedit/blob/HEAD/traning/train_sd3_pix2pix.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"74ab2d24d15218ee"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}