{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UQXSVEZEIIPRE4RDPOTONPGMCZ","short_pith_number":"pith:UQXSVEZE","schema_version":"1.0","canonical_sha256":"a42f2a9324421f1272237ba6e6bccc167053eb99b487ffffe4eef11a011e6b1f","source":{"kind":"arxiv","id":"2504.21292","version":1},"attestation_state":"computed","paper":{"title":"Can We Achieve Efficient Diffusion without Self-Attention? Distilling Self-Attention into Convolutions","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chengxing Zhou, Liang Lin, Pengxu Wei, Weijian Deng, Xiangyang Ji, ZiYi Dong","submitted_at":"2025-04-30T03:57:28Z","abstract_excerpt":"Contemporary diffusion models built upon U-Net or Diffusion Transformer (DiT) architectures have revolutionized image generation through transformer-based attention mechanisms. The prevailing paradigm has commonly employed self-attention with quadratic computational complexity to handle global spatial relationships in complex images, thereby synthesizing high-fidelity images with coherent visual semantics.Contrary to conventional wisdom, our systematic layer-wise analysis reveals an interesting discrepancy: self-attention in pre-trained diffusion models predominantly exhibits localized attenti"},"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":"2504.21292","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-04-30T03:57:28Z","cross_cats_sorted":[],"title_canon_sha256":"2c597607e0ee5bc58179f6611cf4cc17d13ee151b6a8880beff8e3743b7eeb66","abstract_canon_sha256":"f78984ba277df081c376fd21b0721659c69af6e7ccee9b93c17d1d7bf2455d74"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:56:21.639533Z","signature_b64":"qY0TmyUjaZsw1NdhgJxZa/vcUNJetXAmyV4y+f6i+jtYkBFYOYyVOCax3lMd/O0Q+uOBbppQ8vHTGo/CiEW+Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a42f2a9324421f1272237ba6e6bccc167053eb99b487ffffe4eef11a011e6b1f","last_reissued_at":"2026-07-05T10:56:21.639047Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:56:21.639047Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can We Achieve Efficient Diffusion without Self-Attention? Distilling Self-Attention into Convolutions","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chengxing Zhou, Liang Lin, Pengxu Wei, Weijian Deng, Xiangyang Ji, ZiYi Dong","submitted_at":"2025-04-30T03:57:28Z","abstract_excerpt":"Contemporary diffusion models built upon U-Net or Diffusion Transformer (DiT) architectures have revolutionized image generation through transformer-based attention mechanisms. The prevailing paradigm has commonly employed self-attention with quadratic computational complexity to handle global spatial relationships in complex images, thereby synthesizing high-fidelity images with coherent visual semantics.Contrary to conventional wisdom, our systematic layer-wise analysis reveals an interesting discrepancy: self-attention in pre-trained diffusion models predominantly exhibits localized attenti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.21292","kind":"arxiv","version":1},"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/2504.21292/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":"2504.21292","created_at":"2026-07-05T10:56:21.639107+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.21292v1","created_at":"2026-07-05T10:56:21.639107+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.21292","created_at":"2026-07-05T10:56:21.639107+00:00"},{"alias_kind":"pith_short_12","alias_value":"UQXSVEZEIIPR","created_at":"2026-07-05T10:56:21.639107+00:00"},{"alias_kind":"pith_short_16","alias_value":"UQXSVEZEIIPRE4RD","created_at":"2026-07-05T10:56:21.639107+00:00"},{"alias_kind":"pith_short_8","alias_value":"UQXSVEZE","created_at":"2026-07-05T10:56:21.639107+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ","json":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ.json","graph_json":"https://pith.science/api/pith-number/UQXSVEZEIIPRE4RDPOTONPGMCZ/graph.json","events_json":"https://pith.science/api/pith-number/UQXSVEZEIIPRE4RDPOTONPGMCZ/events.json","paper":"https://pith.science/paper/UQXSVEZE"},"agent_actions":{"view_html":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ","download_json":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ.json","view_paper":"https://pith.science/paper/UQXSVEZE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.21292&json=true","fetch_graph":"https://pith.science/api/pith-number/UQXSVEZEIIPRE4RDPOTONPGMCZ/graph.json","fetch_events":"https://pith.science/api/pith-number/UQXSVEZEIIPRE4RDPOTONPGMCZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ/action/storage_attestation","attest_author":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ/action/author_attestation","sign_citation":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ/action/citation_signature","submit_replication":"https://pith.science/pith/UQXSVEZEIIPRE4RDPOTONPGMCZ/action/replication_record"}},"created_at":"2026-07-05T10:56:21.639107+00:00","updated_at":"2026-07-05T10:56:21.639107+00:00"}