{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VO4XBESEGKGR2HIE7CL7KANSYG","short_pith_number":"pith:VO4XBESE","schema_version":"1.0","canonical_sha256":"abb9709244328d1d1d04f897f501b2c1a38ed1c89d86ed96cbc60fb5e38dcaef","source":{"kind":"arxiv","id":"2501.02487","version":3},"attestation_state":"computed","paper":{"title":"ACE++: Instruction-Based Image Creation and Editing via Context-Aware Content Filling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chaojie Mao, Jingfeng Zhang, Jingren Zhou, Yulin Pan, Yu Liu, Zeyinzi Jiang, Zhen Han","submitted_at":"2025-01-05T09:40:58Z","abstract_excerpt":"We report ACE++, an instruction-based diffusion framework that tackles various image generation and editing tasks. Inspired by the input format for the inpainting task proposed by FLUX.1-Fill-dev, we improve the Long-context Condition Unit (LCU) introduced in ACE and extend this input paradigm to any editing and generation tasks. To take full advantage of image generative priors, we develop a two-stage training scheme to minimize the efforts of finetuning powerful text-to-image diffusion models like FLUX.1-dev. In the first stage, we pre-train the model using task data with the 0-ref tasks fro"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2501.02487","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-01-05T09:40:58Z","cross_cats_sorted":[],"title_canon_sha256":"009eb87619d0dfc3150517ac510a4b16671839cbba3242468277f286907f7847","abstract_canon_sha256":"05fb417eb8199b924d9455f6ab69ec9179681a07b626be602479b9a8b22b352d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:01:19.063172Z","signature_b64":"B8Z8hdfXWwPuRjvy1CQNnmkJItvP9CIPixajvZrzw/B3uGDPNoY9b9d5WtI1Nn6pJk5tC52f7pmJKHl0lU4NBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abb9709244328d1d1d04f897f501b2c1a38ed1c89d86ed96cbc60fb5e38dcaef","last_reissued_at":"2026-07-05T10:01:19.062681Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:01:19.062681Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ACE++: Instruction-Based Image Creation and Editing via Context-Aware Content Filling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chaojie Mao, Jingfeng Zhang, Jingren Zhou, Yulin Pan, Yu Liu, Zeyinzi Jiang, Zhen Han","submitted_at":"2025-01-05T09:40:58Z","abstract_excerpt":"We report ACE++, an instruction-based diffusion framework that tackles various image generation and editing tasks. Inspired by the input format for the inpainting task proposed by FLUX.1-Fill-dev, we improve the Long-context Condition Unit (LCU) introduced in ACE and extend this input paradigm to any editing and generation tasks. To take full advantage of image generative priors, we develop a two-stage training scheme to minimize the efforts of finetuning powerful text-to-image diffusion models like FLUX.1-dev. In the first stage, we pre-train the model using task data with the 0-ref tasks fro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02487","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2501.02487/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2501.02487","created_at":"2026-07-05T10:01:19.062747+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.02487v3","created_at":"2026-07-05T10:01:19.062747+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02487","created_at":"2026-07-05T10:01:19.062747+00:00"},{"alias_kind":"pith_short_12","alias_value":"VO4XBESEGKGR","created_at":"2026-07-05T10:01:19.062747+00:00"},{"alias_kind":"pith_short_16","alias_value":"VO4XBESEGKGR2HIE","created_at":"2026-07-05T10:01:19.062747+00:00"},{"alias_kind":"pith_short_8","alias_value":"VO4XBESE","created_at":"2026-07-05T10:01:19.062747+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05071","citing_title":"InstantRetouch: Efficient and High-Fidelity Instruction-Guided Image Retouching with Bilateral Space","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2503.20314","citing_title":"Wan: Open and Advanced Large-Scale Video Generative Models","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17294","citing_title":"HierEdit: Region-Aware Hierarchical Diffusion for Efficient High-Resolution Editing","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2506.18871","citing_title":"OmniGen2: Towards Instruction-Aligned Multimodal Generation","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2504.20690","citing_title":"In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformer","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2503.07598","citing_title":"VACE: All-in-One Video Creation and Editing","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2603.02210","citing_title":"HiFi-Inpaint: Towards High-Fidelity Reference-Based Inpainting for Generating Detail-Preserving Human-Product Images","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10127","citing_title":"Fashion130K: An E-commerce Fashion Dataset for Outfit Generation with Unified Multi-modal Condition","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10127","citing_title":"Fashion130K: An E-commerce Fashion Dataset for Outfit Generation with Unified Multi-modal Condition","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2504.17761","citing_title":"Step1X-Edit: A Practical Framework for General Image Editing","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07457","citing_title":"EditRefiner: A Human-Aligned Agentic Framework for Image Editing Refinement","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17195","citing_title":"DreamShot: Personalized Storyboard Synthesis with Video Diffusion Prior","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19406","citing_title":"HP-Edit: A Human-Preference Post-Training Framework for Image Editing","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG","json":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG.json","graph_json":"https://pith.science/api/pith-number/VO4XBESEGKGR2HIE7CL7KANSYG/graph.json","events_json":"https://pith.science/api/pith-number/VO4XBESEGKGR2HIE7CL7KANSYG/events.json","paper":"https://pith.science/paper/VO4XBESE"},"agent_actions":{"view_html":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG","download_json":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG.json","view_paper":"https://pith.science/paper/VO4XBESE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.02487&json=true","fetch_graph":"https://pith.science/api/pith-number/VO4XBESEGKGR2HIE7CL7KANSYG/graph.json","fetch_events":"https://pith.science/api/pith-number/VO4XBESEGKGR2HIE7CL7KANSYG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG/action/storage_attestation","attest_author":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG/action/author_attestation","sign_citation":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG/action/citation_signature","submit_replication":"https://pith.science/pith/VO4XBESEGKGR2HIE7CL7KANSYG/action/replication_record"}},"created_at":"2026-07-05T10:01:19.062747+00:00","updated_at":"2026-07-05T10:01:19.062747+00:00"}