{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WKL3INGZHWK2JZH7TQVA7H4ESR","short_pith_number":"pith:WKL3INGZ","schema_version":"1.0","canonical_sha256":"b297b434d93d95a4e4ff9c2a0f9f849462e7d96d58922abaed931182a6eed231","source":{"kind":"arxiv","id":"2305.15194","version":3},"attestation_state":"computed","paper":{"title":"DiffBlender: Composable and Versatile Multimodal Text-to-Image Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Daesik Kim, Junsoo Lee, Kibeom Hong, Namhyuk Ahn, Sungnyun Kim","submitted_at":"2023-05-24T14:31:20Z","abstract_excerpt":"In this study, we aim to enhance the capabilities of diffusion-based text-to-image (T2I) generation models by integrating diverse modalities beyond textual descriptions within a unified framework. To this end, we categorize widely used conditional inputs into three modality types: structure, layout, and attribute. We propose a multimodal T2I diffusion model, which is capable of processing all three modalities within a single architecture without modifying the parameters of the pre-trained diffusion model, as only a small subset of components is updated. Our approach sets new benchmarks in mult"},"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":"2305.15194","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-24T14:31:20Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6838346db7fa54a10c12e72f42f39bdcba8be8a0d6212c41dfa1b2e8824cd3de","abstract_canon_sha256":"83e6453c6684ea4e9d24626ba9f9de7a76bcd3dca04f639c7eebf74e7dc9a0dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:03.077949Z","signature_b64":"nzpvOtfm79IaWgl85obUru5Kywn8vSN3K8O8jnjTaqAztaf0LobhiGpnN20Zo9rQDc9dcHh9WJ2R65Hl5BXgCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b297b434d93d95a4e4ff9c2a0f9f849462e7d96d58922abaed931182a6eed231","last_reissued_at":"2026-07-05T11:59:03.077494Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:03.077494Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DiffBlender: Composable and Versatile Multimodal Text-to-Image Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Daesik Kim, Junsoo Lee, Kibeom Hong, Namhyuk Ahn, Sungnyun Kim","submitted_at":"2023-05-24T14:31:20Z","abstract_excerpt":"In this study, we aim to enhance the capabilities of diffusion-based text-to-image (T2I) generation models by integrating diverse modalities beyond textual descriptions within a unified framework. To this end, we categorize widely used conditional inputs into three modality types: structure, layout, and attribute. We propose a multimodal T2I diffusion model, which is capable of processing all three modalities within a single architecture without modifying the parameters of the pre-trained diffusion model, as only a small subset of components is updated. Our approach sets new benchmarks in mult"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.15194","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/2305.15194/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":"2305.15194","created_at":"2026-07-05T11:59:03.077549+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.15194v3","created_at":"2026-07-05T11:59:03.077549+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.15194","created_at":"2026-07-05T11:59:03.077549+00:00"},{"alias_kind":"pith_short_12","alias_value":"WKL3INGZHWK2","created_at":"2026-07-05T11:59:03.077549+00:00"},{"alias_kind":"pith_short_16","alias_value":"WKL3INGZHWK2JZH7","created_at":"2026-07-05T11:59:03.077549+00:00"},{"alias_kind":"pith_short_8","alias_value":"WKL3INGZ","created_at":"2026-07-05T11:59:03.077549+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2405.18716","citing_title":"SketchDeco: Training-Free Latent Composition for Precise Sketch Colourisation","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2603.14209","citing_title":"ChArtist: Generating Pictorial Charts with Unified Spatial and Subject Control","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR","json":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR.json","graph_json":"https://pith.science/api/pith-number/WKL3INGZHWK2JZH7TQVA7H4ESR/graph.json","events_json":"https://pith.science/api/pith-number/WKL3INGZHWK2JZH7TQVA7H4ESR/events.json","paper":"https://pith.science/paper/WKL3INGZ"},"agent_actions":{"view_html":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR","download_json":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR.json","view_paper":"https://pith.science/paper/WKL3INGZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.15194&json=true","fetch_graph":"https://pith.science/api/pith-number/WKL3INGZHWK2JZH7TQVA7H4ESR/graph.json","fetch_events":"https://pith.science/api/pith-number/WKL3INGZHWK2JZH7TQVA7H4ESR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR/action/storage_attestation","attest_author":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR/action/author_attestation","sign_citation":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR/action/citation_signature","submit_replication":"https://pith.science/pith/WKL3INGZHWK2JZH7TQVA7H4ESR/action/replication_record"}},"created_at":"2026-07-05T11:59:03.077549+00:00","updated_at":"2026-07-05T11:59:03.077549+00:00"}