{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OUPFA43AXQLFZNIKFGCKEMMNET","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6eebd63656fe36dd4ba1207d47286fb487a5e31f45674d3556b202bb3bb528e4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-28T18:00:12Z","title_canon_sha256":"f2325950a49bb3ec4debbe596076ea06255a9e50262dfd37b8d7c4b48f113fff"},"schema_version":"1.0","source":{"id":"2503.22796","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.22796","created_at":"2026-07-05T10:41:25Z"},{"alias_kind":"arxiv_version","alias_value":"2503.22796v1","created_at":"2026-07-05T10:41:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.22796","created_at":"2026-07-05T10:41:25Z"},{"alias_kind":"pith_short_12","alias_value":"OUPFA43AXQLF","created_at":"2026-07-05T10:41:25Z"},{"alias_kind":"pith_short_16","alias_value":"OUPFA43AXQLFZNIK","created_at":"2026-07-05T10:41:25Z"},{"alias_kind":"pith_short_8","alias_value":"OUPFA43A","created_at":"2026-07-05T10:41:25Z"}],"graph_snapshots":[{"event_id":"sha256:6a1f10ffeddd710161542b82efa7d275960dca8c13734b8a50b74a6562dad00a","target":"graph","created_at":"2026-07-05T10:41:25Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2503.22796/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Text-to-image generation models, especially Multimodal Diffusion Transformers (MMDiT), have shown remarkable progress in generating high-quality images. However, these models often face significant computational bottlenecks, particularly in attention mechanisms, which hinder their scalability and efficiency. In this paper, we introduce DiTFastAttnV2, a post-training compression method designed to accelerate attention in MMDiT. Through an in-depth analysis of MMDiT's attention patterns, we identify key differences from prior DiT-based methods and propose head-wise arrow attention and caching me","authors_text":"Guohao Dai, Hanling Zhang, Mingzhu Shen Yibo Fan, Pengtao Chen, Rundong Su, Shengen Yan, Yu Wang, Zhihang Yuan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-28T18:00:12Z","title":"DiTFastAttnV2: Head-wise Attention Compression for Multi-Modality Diffusion Transformers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.22796","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:fd98a8af43475c3ed0e35d0e6d3d68d4c79ac56dc2827b2c64e501090d0c715d","target":"record","created_at":"2026-07-05T10:41:25Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6eebd63656fe36dd4ba1207d47286fb487a5e31f45674d3556b202bb3bb528e4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-28T18:00:12Z","title_canon_sha256":"f2325950a49bb3ec4debbe596076ea06255a9e50262dfd37b8d7c4b48f113fff"},"schema_version":"1.0","source":{"id":"2503.22796","kind":"arxiv","version":1}},"canonical_sha256":"751e507360bc165cb50a2984a2318d24ec32e4533c2597bcf15025eebd0b1e0b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"751e507360bc165cb50a2984a2318d24ec32e4533c2597bcf15025eebd0b1e0b","first_computed_at":"2026-07-05T10:41:25.947231Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:41:25.947231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dLWxbSe46x1140PuV37ut9755AHZr7znMGzbZU6XscCtvA7vAdLPPne2SHqONBWBYyEoQFTXIxT1n5baCJk/AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:41:25.947724Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.22796","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fd98a8af43475c3ed0e35d0e6d3d68d4c79ac56dc2827b2c64e501090d0c715d","sha256:6a1f10ffeddd710161542b82efa7d275960dca8c13734b8a50b74a6562dad00a"],"state_sha256":"aa33cf931ac528c05cb717e72a50acce09e06a398e717961c2c22ff30e508484"}