{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:44JARK7SWY3VWIVQNN7YND3PMI","short_pith_number":"pith:44JARK7S","schema_version":"1.0","canonical_sha256":"e71208abf2b6375b22b06b7f868f6f6200c20dacedc6326d1535a260223a552b","source":{"kind":"arxiv","id":"2404.07987","version":4},"attestation_state":"computed","paper":{"title":"ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Chen Chen, Huafeng Kuang, Jie Wu, Ming Li, Taojiannan Yang, Xuefeng Xiao, Zhaoning Wang","submitted_at":"2024-04-11T17:59:09Z","abstract_excerpt":"To enhance the controllability of text-to-image diffusion models, existing efforts like ControlNet incorporated image-based conditional controls. In this paper, we reveal that existing methods still face significant challenges in generating images that align with the image conditional controls. To this end, we propose ControlNet++, a novel approach that improves controllable generation by explicitly optimizing pixel-level cycle consistency between generated images and conditional controls. Specifically, for an input conditional control, we use a pre-trained discriminative reward model to extra"},"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":"2404.07987","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-04-11T17:59:09Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"458e4e91c580aebb4ca1131352f10e082773fdc69800911c37a311785a094900","abstract_canon_sha256":"9473c3b4cc9f16f6416c8a5fda007ee0ca4c2660b717cbaa9cffa16502a39bb3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:37:07.387209Z","signature_b64":"iZHbFrb2qDGOCuFv4fLn3K7Q7F94Dkh7HHahFB92LGForDDT5DrETD6hIn0iKCL6ZxVmV813I7vvgKL1cLjlCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e71208abf2b6375b22b06b7f868f6f6200c20dacedc6326d1535a260223a552b","last_reissued_at":"2026-07-05T09:37:07.386589Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:37:07.386589Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Chen Chen, Huafeng Kuang, Jie Wu, Ming Li, Taojiannan Yang, Xuefeng Xiao, Zhaoning Wang","submitted_at":"2024-04-11T17:59:09Z","abstract_excerpt":"To enhance the controllability of text-to-image diffusion models, existing efforts like ControlNet incorporated image-based conditional controls. In this paper, we reveal that existing methods still face significant challenges in generating images that align with the image conditional controls. To this end, we propose ControlNet++, a novel approach that improves controllable generation by explicitly optimizing pixel-level cycle consistency between generated images and conditional controls. Specifically, for an input conditional control, we use a pre-trained discriminative reward model to extra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.07987","kind":"arxiv","version":4},"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/2404.07987/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":"2404.07987","created_at":"2026-07-05T09:37:07.386671+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.07987v4","created_at":"2026-07-05T09:37:07.386671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.07987","created_at":"2026-07-05T09:37:07.386671+00:00"},{"alias_kind":"pith_short_12","alias_value":"44JARK7SWY3V","created_at":"2026-07-05T09:37:07.386671+00:00"},{"alias_kind":"pith_short_16","alias_value":"44JARK7SWY3VWIVQ","created_at":"2026-07-05T09:37:07.386671+00:00"},{"alias_kind":"pith_short_8","alias_value":"44JARK7S","created_at":"2026-07-05T09:37:07.386671+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.06608","citing_title":"TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow Models","ref_index":135,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06870","citing_title":"RefineAnything: Multimodal Region-Specific Refinement for Perfect Local Details","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI","json":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI.json","graph_json":"https://pith.science/api/pith-number/44JARK7SWY3VWIVQNN7YND3PMI/graph.json","events_json":"https://pith.science/api/pith-number/44JARK7SWY3VWIVQNN7YND3PMI/events.json","paper":"https://pith.science/paper/44JARK7S"},"agent_actions":{"view_html":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI","download_json":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI.json","view_paper":"https://pith.science/paper/44JARK7S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.07987&json=true","fetch_graph":"https://pith.science/api/pith-number/44JARK7SWY3VWIVQNN7YND3PMI/graph.json","fetch_events":"https://pith.science/api/pith-number/44JARK7SWY3VWIVQNN7YND3PMI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI/action/storage_attestation","attest_author":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI/action/author_attestation","sign_citation":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI/action/citation_signature","submit_replication":"https://pith.science/pith/44JARK7SWY3VWIVQNN7YND3PMI/action/replication_record"}},"created_at":"2026-07-05T09:37:07.386671+00:00","updated_at":"2026-07-05T09:37:07.386671+00:00"}