{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:33PF7K2IGWTYZZVKX3GRTPDHN5","short_pith_number":"pith:33PF7K2I","schema_version":"1.0","canonical_sha256":"dede5fab4835a78ce6aabecd19bc676f4cdd79b19aef40b82eb528e89a10958f","source":{"kind":"arxiv","id":"2302.12764","version":2},"attestation_state":"computed","paper":{"title":"Modulating Pretrained Diffusion Models for Multimodal Image Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cusuh Ham, James Hays, Jingwan Lu, Krishna Kumar Singh, Tobias Hinz, Zhifei Zhang","submitted_at":"2023-02-24T17:28:08Z","abstract_excerpt":"We present multimodal conditioning modules (MCM) for enabling conditional image synthesis using pretrained diffusion models. Previous multimodal synthesis works rely on training networks from scratch or fine-tuning pretrained networks, both of which are computationally expensive for large, state-of-the-art diffusion models. Our method uses pretrained networks but \\textit{does not require any updates to the diffusion network's parameters}. MCM is a small module trained to modulate the diffusion network's predictions during sampling using 2D modalities (e.g., semantic segmentation maps, sketches"},"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":"2302.12764","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-02-24T17:28:08Z","cross_cats_sorted":[],"title_canon_sha256":"cf8a21b6e98100d8be3dd11974511a94f23e26f40f8640112fa33ac18002e8f6","abstract_canon_sha256":"aa59867133cdd0f72d4a465233c171498d3de6def42efe89273df88464e0adc8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:11:32.692293Z","signature_b64":"YcEoaU6/q2q4O8gM8XsALwcX65urzyOG0JIUOtIwdtiNCDHtTNokQfwVCCpskoINsk/9DA7TGUR0SSCzVwZYAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dede5fab4835a78ce6aabecd19bc676f4cdd79b19aef40b82eb528e89a10958f","last_reissued_at":"2026-07-05T06:11:32.691829Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:11:32.691829Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modulating Pretrained Diffusion Models for Multimodal Image Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cusuh Ham, James Hays, Jingwan Lu, Krishna Kumar Singh, Tobias Hinz, Zhifei Zhang","submitted_at":"2023-02-24T17:28:08Z","abstract_excerpt":"We present multimodal conditioning modules (MCM) for enabling conditional image synthesis using pretrained diffusion models. Previous multimodal synthesis works rely on training networks from scratch or fine-tuning pretrained networks, both of which are computationally expensive for large, state-of-the-art diffusion models. Our method uses pretrained networks but \\textit{does not require any updates to the diffusion network's parameters}. MCM is a small module trained to modulate the diffusion network's predictions during sampling using 2D modalities (e.g., semantic segmentation maps, sketches"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.12764","kind":"arxiv","version":2},"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/2302.12764/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":"2302.12764","created_at":"2026-07-05T06:11:32.691888+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.12764v2","created_at":"2026-07-05T06:11:32.691888+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.12764","created_at":"2026-07-05T06:11:32.691888+00:00"},{"alias_kind":"pith_short_12","alias_value":"33PF7K2IGWTY","created_at":"2026-07-05T06:11:32.691888+00:00"},{"alias_kind":"pith_short_16","alias_value":"33PF7K2IGWTYZZVK","created_at":"2026-07-05T06:11:32.691888+00:00"},{"alias_kind":"pith_short_8","alias_value":"33PF7K2I","created_at":"2026-07-05T06:11:32.691888+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.05501","citing_title":"Preliminary Explorations with GPT-4o(mni) Native Image Generation","ref_index":51,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5","json":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5.json","graph_json":"https://pith.science/api/pith-number/33PF7K2IGWTYZZVKX3GRTPDHN5/graph.json","events_json":"https://pith.science/api/pith-number/33PF7K2IGWTYZZVKX3GRTPDHN5/events.json","paper":"https://pith.science/paper/33PF7K2I"},"agent_actions":{"view_html":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5","download_json":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5.json","view_paper":"https://pith.science/paper/33PF7K2I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.12764&json=true","fetch_graph":"https://pith.science/api/pith-number/33PF7K2IGWTYZZVKX3GRTPDHN5/graph.json","fetch_events":"https://pith.science/api/pith-number/33PF7K2IGWTYZZVKX3GRTPDHN5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5/action/storage_attestation","attest_author":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5/action/author_attestation","sign_citation":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5/action/citation_signature","submit_replication":"https://pith.science/pith/33PF7K2IGWTYZZVKX3GRTPDHN5/action/replication_record"}},"created_at":"2026-07-05T06:11:32.691888+00:00","updated_at":"2026-07-05T06:11:32.691888+00:00"}