{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FTN76RF6UUB2HPMKDVM6US76DF","short_pith_number":"pith:FTN76RF6","schema_version":"1.0","canonical_sha256":"2cdbff44bea503a3bd8a1d59ea4bfe1959eec6c279275e36fab9fd07f72993d2","source":{"kind":"arxiv","id":"2503.19881","version":1},"attestation_state":"computed","paper":{"title":"Mask$^2$DiT: Dual Mask-based Diffusion Transformer for Multi-Scene Long Video Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongtao Xie, Jianlong Yuan, Jiawei Liu, Qian He, Shancheng Fang, Siyu Zhou, Tianhao Qi, Wanquan Feng, Yongdong Zhang","submitted_at":"2025-03-25T17:46:50Z","abstract_excerpt":"Sora has unveiled the immense potential of the Diffusion Transformer (DiT) architecture in single-scene video generation. However, the more challenging task of multi-scene video generation, which offers broader applications, remains relatively underexplored. To bridge this gap, we propose Mask$^2$DiT, a novel approach that establishes fine-grained, one-to-one alignment between video segments and their corresponding text annotations. Specifically, we introduce a symmetric binary mask at each attention layer within the DiT architecture, ensuring that each text annotation applies exclusively to i"},"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":"2503.19881","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-03-25T17:46:50Z","cross_cats_sorted":[],"title_canon_sha256":"81998303a90a18602222de32b065a0b65b659cc9aa87f45944be3a6763cde8f2","abstract_canon_sha256":"97942aae2713572fb1f3747f616cd54ee52442b81ab3c77d46dc7a3d8397ef26"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:39:03.055557Z","signature_b64":"tQnt1QsvM89tIw4URzAzK7k2TwW08tViheB/6SWX362CJhLQmywuTeiI/LWVbOgGGvtWRMaNMGFXkGaE860WBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cdbff44bea503a3bd8a1d59ea4bfe1959eec6c279275e36fab9fd07f72993d2","last_reissued_at":"2026-07-05T10:39:03.055036Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:39:03.055036Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mask$^2$DiT: Dual Mask-based Diffusion Transformer for Multi-Scene Long Video Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongtao Xie, Jianlong Yuan, Jiawei Liu, Qian He, Shancheng Fang, Siyu Zhou, Tianhao Qi, Wanquan Feng, Yongdong Zhang","submitted_at":"2025-03-25T17:46:50Z","abstract_excerpt":"Sora has unveiled the immense potential of the Diffusion Transformer (DiT) architecture in single-scene video generation. However, the more challenging task of multi-scene video generation, which offers broader applications, remains relatively underexplored. To bridge this gap, we propose Mask$^2$DiT, a novel approach that establishes fine-grained, one-to-one alignment between video segments and their corresponding text annotations. Specifically, we introduce a symmetric binary mask at each attention layer within the DiT architecture, ensuring that each text annotation applies exclusively to i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.19881","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/2503.19881/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":"2503.19881","created_at":"2026-07-05T10:39:03.055111+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.19881v1","created_at":"2026-07-05T10:39:03.055111+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.19881","created_at":"2026-07-05T10:39:03.055111+00:00"},{"alias_kind":"pith_short_12","alias_value":"FTN76RF6UUB2","created_at":"2026-07-05T10:39:03.055111+00:00"},{"alias_kind":"pith_short_16","alias_value":"FTN76RF6UUB2HPMK","created_at":"2026-07-05T10:39:03.055111+00:00"},{"alias_kind":"pith_short_8","alias_value":"FTN76RF6","created_at":"2026-07-05T10:39:03.055111+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20799","citing_title":"GroundShot: Visually Consistent Multi-Shot Long Video Generation via Entity-Grounded Shot Scheduling","ref_index":39,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF","json":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF.json","graph_json":"https://pith.science/api/pith-number/FTN76RF6UUB2HPMKDVM6US76DF/graph.json","events_json":"https://pith.science/api/pith-number/FTN76RF6UUB2HPMKDVM6US76DF/events.json","paper":"https://pith.science/paper/FTN76RF6"},"agent_actions":{"view_html":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF","download_json":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF.json","view_paper":"https://pith.science/paper/FTN76RF6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.19881&json=true","fetch_graph":"https://pith.science/api/pith-number/FTN76RF6UUB2HPMKDVM6US76DF/graph.json","fetch_events":"https://pith.science/api/pith-number/FTN76RF6UUB2HPMKDVM6US76DF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF/action/storage_attestation","attest_author":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF/action/author_attestation","sign_citation":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF/action/citation_signature","submit_replication":"https://pith.science/pith/FTN76RF6UUB2HPMKDVM6US76DF/action/replication_record"}},"created_at":"2026-07-05T10:39:03.055111+00:00","updated_at":"2026-07-05T10:39:03.055111+00:00"}