{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DRDHCKGV4O3ZNSNPBATTRGA632","short_pith_number":"pith:DRDHCKGV","schema_version":"1.0","canonical_sha256":"1c467128d5e3b796c9af082738981ede971e94dd5ec158f4dd610743ba56b823","source":{"kind":"arxiv","id":"2606.19365","version":1},"attestation_state":"computed","paper":{"title":"Performance Analysis and Optimization of 3D Generative Diffusion Models across GPU Architectures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Byeong Kil Lee, Jeeho Ryoo, Jiatong Han, Muhammad Ali Khaliq, Weidong Zhang, Yongchan Jung","submitted_at":"2026-06-11T02:12:10Z","abstract_excerpt":"Diffusion models have become essential for high-fidelity 3D MRI synthesis, yet their deployment remains constrained by substantial GPU resource demands arising from hundreds of U-Net evaluations per sample and a highly heterogeneous kernel behavior. This paper performs a comprehensive performance analysis of the state-of-the-art medical diffusion model, Med-DDPM, across three generations of NVIDIA architectures to study kernel-level runtime breakdowns, instruction-mix characteristics, memory system utilization, warp-level activities, and profiler priority-score estimates. We show that training"},"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":"2606.19365","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-11T02:12:10Z","cross_cats_sorted":[],"title_canon_sha256":"71eaa56d910b1d5ae00de6e573e191a0a975484b7e694bf03ef013bc675ad2b9","abstract_canon_sha256":"9066b93a0e4749a7042c83aeabbf6e8752cb52e2eebdb12ddb81452e39f36411"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:23.274642Z","signature_b64":"8XqKnE3rSM1qCbl8rPQQDSOdZHgoOG4+WxHWshe0rquD7aph8NhCDvJxKWtGt6cfHDaMroxpXaLcS0BiRg4vAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1c467128d5e3b796c9af082738981ede971e94dd5ec158f4dd610743ba56b823","last_reissued_at":"2026-06-19T16:12:23.274290Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:23.274290Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Performance Analysis and Optimization of 3D Generative Diffusion Models across GPU Architectures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Byeong Kil Lee, Jeeho Ryoo, Jiatong Han, Muhammad Ali Khaliq, Weidong Zhang, Yongchan Jung","submitted_at":"2026-06-11T02:12:10Z","abstract_excerpt":"Diffusion models have become essential for high-fidelity 3D MRI synthesis, yet their deployment remains constrained by substantial GPU resource demands arising from hundreds of U-Net evaluations per sample and a highly heterogeneous kernel behavior. This paper performs a comprehensive performance analysis of the state-of-the-art medical diffusion model, Med-DDPM, across three generations of NVIDIA architectures to study kernel-level runtime breakdowns, instruction-mix characteristics, memory system utilization, warp-level activities, and profiler priority-score estimates. We show that training"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19365","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/2606.19365/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":"2606.19365","created_at":"2026-06-19T16:12:23.274362+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.19365v1","created_at":"2026-06-19T16:12:23.274362+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19365","created_at":"2026-06-19T16:12:23.274362+00:00"},{"alias_kind":"pith_short_12","alias_value":"DRDHCKGV4O3Z","created_at":"2026-06-19T16:12:23.274362+00:00"},{"alias_kind":"pith_short_16","alias_value":"DRDHCKGV4O3ZNSNP","created_at":"2026-06-19T16:12:23.274362+00:00"},{"alias_kind":"pith_short_8","alias_value":"DRDHCKGV","created_at":"2026-06-19T16:12:23.274362+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/DRDHCKGV4O3ZNSNPBATTRGA632","json":"https://pith.science/pith/DRDHCKGV4O3ZNSNPBATTRGA632.json","graph_json":"https://pith.science/api/pith-number/DRDHCKGV4O3ZNSNPBATTRGA632/graph.json","events_json":"https://pith.science/api/pith-number/DRDHCKGV4O3ZNSNPBATTRGA632/events.json","paper":"https://pith.science/paper/DRDHCKGV"},"agent_actions":{"view_html":"https://pith.science/pith/DRDHCKGV4O3ZNSNPBATTRGA632","download_json":"https://pith.science/pith/DRDHCKGV4O3ZNSNPBATTRGA632.json","view_paper":"https://pith.science/paper/DRDHCKGV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.19365&json=true","fetch_graph":"https://pith.science/api/pith-number/DRDHCKGV4O3ZNSNPBATTRGA632/graph.json","fetch_events":"https://pith.science/api/pith-number/DRDHCKGV4O3ZNSNPBATTRGA632/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DRDHCKGV4O3ZNSNPBATTRGA632/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DRDHCKGV4O3ZNSNPBATTRGA632/action/storage_attestation","attest_author":"https://pith.science/pith/DRDHCKGV4O3ZNSNPBATTRGA632/action/author_attestation","sign_citation":"https://pith.science/pith/DRDHCKGV4O3ZNSNPBATTRGA632/action/citation_signature","submit_replication":"https://pith.science/pith/DRDHCKGV4O3ZNSNPBATTRGA632/action/replication_record"}},"created_at":"2026-06-19T16:12:23.274362+00:00","updated_at":"2026-06-19T16:12:23.274362+00:00"}