{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5RONDWIKEX73YEDCA3OKWQL5R3","short_pith_number":"pith:5RONDWIK","schema_version":"1.0","canonical_sha256":"ec5cd1d90a25ffbc106206dcab417d8ee5eb32326a327076a745428f4f1b2a46","source":{"kind":"arxiv","id":"2405.03958","version":3},"attestation_state":"computed","paper":{"title":"Simple Drop-in LoRA Conditioning on Attention Layers Will Improve Your Diffusion Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Albert No, Ernest K. Ryu, Inkyu Park, Jaesung R. Park, Jaewoong Cho, Joo Young Choi","submitted_at":"2024-05-07T02:45:28Z","abstract_excerpt":"Current state-of-the-art diffusion models employ U-Net architectures containing convolutional and (qkv) self-attention layers. The U-Net processes images while being conditioned on the time embedding input for each sampling step and the class or caption embedding input corresponding to the desired conditional generation. Such conditioning involves scale-and-shift operations to the convolutional layers but does not directly affect the attention layers. While these standard architectural choices are certainly effective, not conditioning the attention layers feels arbitrary and potentially subopt"},"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":"2405.03958","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-07T02:45:28Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"5ec899143cec45ba412a34d9ebfd12748f4a60114881d13525dbbf01a32205df","abstract_canon_sha256":"d14b22a663ddb7107f44aad85909815008b8b4f50e588a3fb7ced3ff2d11fba4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:32.303853Z","signature_b64":"7/bTUIHDD+C6ef8ccvkBtk0+Hw+0sVSlKnW1mE5dGtSwV3ObPozvGBHEGQaUL9u1GorHanNzA4MTn9DBBu3vAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ec5cd1d90a25ffbc106206dcab417d8ee5eb32326a327076a745428f4f1b2a46","last_reissued_at":"2026-07-05T09:15:32.303311Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:32.303311Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Simple Drop-in LoRA Conditioning on Attention Layers Will Improve Your Diffusion Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Albert No, Ernest K. Ryu, Inkyu Park, Jaesung R. Park, Jaewoong Cho, Joo Young Choi","submitted_at":"2024-05-07T02:45:28Z","abstract_excerpt":"Current state-of-the-art diffusion models employ U-Net architectures containing convolutional and (qkv) self-attention layers. The U-Net processes images while being conditioned on the time embedding input for each sampling step and the class or caption embedding input corresponding to the desired conditional generation. Such conditioning involves scale-and-shift operations to the convolutional layers but does not directly affect the attention layers. While these standard architectural choices are certainly effective, not conditioning the attention layers feels arbitrary and potentially subopt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.03958","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/2405.03958/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":"2405.03958","created_at":"2026-07-05T09:15:32.303376+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.03958v3","created_at":"2026-07-05T09:15:32.303376+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.03958","created_at":"2026-07-05T09:15:32.303376+00:00"},{"alias_kind":"pith_short_12","alias_value":"5RONDWIKEX73","created_at":"2026-07-05T09:15:32.303376+00:00"},{"alias_kind":"pith_short_16","alias_value":"5RONDWIKEX73YEDC","created_at":"2026-07-05T09:15:32.303376+00:00"},{"alias_kind":"pith_short_8","alias_value":"5RONDWIK","created_at":"2026-07-05T09:15:32.303376+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.05063","citing_title":"CytoDiff: AI-Driven Cytomorphology Image Synthesis for Medical Diagnostics","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3","json":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3.json","graph_json":"https://pith.science/api/pith-number/5RONDWIKEX73YEDCA3OKWQL5R3/graph.json","events_json":"https://pith.science/api/pith-number/5RONDWIKEX73YEDCA3OKWQL5R3/events.json","paper":"https://pith.science/paper/5RONDWIK"},"agent_actions":{"view_html":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3","download_json":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3.json","view_paper":"https://pith.science/paper/5RONDWIK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.03958&json=true","fetch_graph":"https://pith.science/api/pith-number/5RONDWIKEX73YEDCA3OKWQL5R3/graph.json","fetch_events":"https://pith.science/api/pith-number/5RONDWIKEX73YEDCA3OKWQL5R3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3/action/storage_attestation","attest_author":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3/action/author_attestation","sign_citation":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3/action/citation_signature","submit_replication":"https://pith.science/pith/5RONDWIKEX73YEDCA3OKWQL5R3/action/replication_record"}},"created_at":"2026-07-05T09:15:32.303376+00:00","updated_at":"2026-07-05T09:15:32.303376+00:00"}