{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CNQ5WVKLIEM6UIXBAOU6N2RFBS","short_pith_number":"pith:CNQ5WVKL","schema_version":"1.0","canonical_sha256":"1361db554b4119ea22e103a9e6ea250c84bf78a7334512da13f39be272dc043f","source":{"kind":"arxiv","id":"2608.03335","version":1},"attestation_state":"computed","paper":{"title":"SPADE: An Input-Adaptive Sparse Attention Engine for Fast Video Diffusion Models Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hailong Yang, Renze Chen, Shanghao Liu, Size Zheng, Yuanqiang Liu, Yun (Eric) Liang","submitted_at":"2026-08-04T08:47:57Z","abstract_excerpt":"Video diffusion transformers (vDiTs) generate high quality but pay quadratic self-attention cost, making inference prohibitive at video-token scales. The challenge is input-adaptive sparsity: selecting critical Q/K/V tokens with negligible overhead and executing them for end-to-end gains. We present SPADE, a training-free sparse-attention engine of three parts: (i) vDiT-SSR, a specification defining 3D blocking candidates and formalizing dynamic masks via Summarizer/Estimator expressions; (ii) runtime scheme generation using SICS and a head-wise policy; and (iii) an executor with low-overhead "},"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":"2608.03335","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-04T08:47:57Z","cross_cats_sorted":[],"title_canon_sha256":"bdac8fac6bb9e22663884a30a1b5f5b6c313ee7ae5a8901a4b6b3567639464fa","abstract_canon_sha256":"02544443da1fc9ebdd47b9bea9ac285507aee90dffd03fca79d2fc9bd651848d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T00:46:55.843179Z","signature_b64":"yCXdsMp/VWlaQUog4iWCLL5bueF0Xjmgz5jwJfZz/JaP9Ul91r6wEcaRTZpPVkHObA9/Nv5/6fiYDQvo0m0ADQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1361db554b4119ea22e103a9e6ea250c84bf78a7334512da13f39be272dc043f","last_reissued_at":"2026-08-05T00:46:55.839187Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T00:46:55.839187Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SPADE: An Input-Adaptive Sparse Attention Engine for Fast Video Diffusion Models Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hailong Yang, Renze Chen, Shanghao Liu, Size Zheng, Yuanqiang Liu, Yun (Eric) Liang","submitted_at":"2026-08-04T08:47:57Z","abstract_excerpt":"Video diffusion transformers (vDiTs) generate high quality but pay quadratic self-attention cost, making inference prohibitive at video-token scales. The challenge is input-adaptive sparsity: selecting critical Q/K/V tokens with negligible overhead and executing them for end-to-end gains. We present SPADE, a training-free sparse-attention engine of three parts: (i) vDiT-SSR, a specification defining 3D blocking candidates and formalizing dynamic masks via Summarizer/Estimator expressions; (ii) runtime scheme generation using SICS and a head-wise policy; and (iii) an executor with low-overhead "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03335","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/2608.03335/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":"2608.03335","created_at":"2026-08-05T00:46:55.840184+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.03335v1","created_at":"2026-08-05T00:46:55.840184+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03335","created_at":"2026-08-05T00:46:55.840184+00:00"},{"alias_kind":"pith_short_12","alias_value":"CNQ5WVKLIEM6","created_at":"2026-08-05T00:46:55.840184+00:00"},{"alias_kind":"pith_short_16","alias_value":"CNQ5WVKLIEM6UIXB","created_at":"2026-08-05T00:46:55.840184+00:00"},{"alias_kind":"pith_short_8","alias_value":"CNQ5WVKL","created_at":"2026-08-05T00:46:55.840184+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/CNQ5WVKLIEM6UIXBAOU6N2RFBS","json":"https://pith.science/pith/CNQ5WVKLIEM6UIXBAOU6N2RFBS.json","graph_json":"https://pith.science/api/pith-number/CNQ5WVKLIEM6UIXBAOU6N2RFBS/graph.json","events_json":"https://pith.science/api/pith-number/CNQ5WVKLIEM6UIXBAOU6N2RFBS/events.json","paper":"https://pith.science/paper/CNQ5WVKL"},"agent_actions":{"view_html":"https://pith.science/pith/CNQ5WVKLIEM6UIXBAOU6N2RFBS","download_json":"https://pith.science/pith/CNQ5WVKLIEM6UIXBAOU6N2RFBS.json","view_paper":"https://pith.science/paper/CNQ5WVKL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.03335&json=true","fetch_graph":"https://pith.science/api/pith-number/CNQ5WVKLIEM6UIXBAOU6N2RFBS/graph.json","fetch_events":"https://pith.science/api/pith-number/CNQ5WVKLIEM6UIXBAOU6N2RFBS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CNQ5WVKLIEM6UIXBAOU6N2RFBS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CNQ5WVKLIEM6UIXBAOU6N2RFBS/action/storage_attestation","attest_author":"https://pith.science/pith/CNQ5WVKLIEM6UIXBAOU6N2RFBS/action/author_attestation","sign_citation":"https://pith.science/pith/CNQ5WVKLIEM6UIXBAOU6N2RFBS/action/citation_signature","submit_replication":"https://pith.science/pith/CNQ5WVKLIEM6UIXBAOU6N2RFBS/action/replication_record"}},"created_at":"2026-08-05T00:46:55.840184+00:00","updated_at":"2026-08-05T00:46:55.840184+00:00"}