{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:P4DTAZY6KSBOHN5NSSOWEEV6IW","short_pith_number":"pith:P4DTAZY6","schema_version":"1.0","canonical_sha256":"7f0730671e5482e3b7ad949d6212be45af54245bf836e839cf4cc83bf247bfd1","source":{"kind":"arxiv","id":"2507.12952","version":1},"attestation_state":"computed","paper":{"title":"LoViC: Efficient Long Video Generation with Context Compression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Han Wu, Jiaxiu Jiang, Jingjing Ren, Wangmeng Zuo, Wenbo Li, Xiaogang Xu, Yong Guo, Yuping Qiu","submitted_at":"2025-07-17T09:46:43Z","abstract_excerpt":"Despite recent advances in diffusion transformers (DiTs) for text-to-video generation, scaling to long-duration content remains challenging due to the quadratic complexity of self-attention. While prior efforts -- such as sparse attention and temporally autoregressive models -- offer partial relief, they often compromise temporal coherence or scalability. We introduce LoViC, a DiT-based framework trained on million-scale open-domain videos, designed to produce long, coherent videos through a segment-wise generation process. At the core of our approach is FlexFormer, an expressive autoencoder t"},"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":"2507.12952","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-17T09:46:43Z","cross_cats_sorted":[],"title_canon_sha256":"f545486a983ac3c1a2b971aa56640594c0e9cf51e3b11d95aac7bd5483c33867","abstract_canon_sha256":"641ac23505282090d1dacad85dc4fac060e4da1d128a0973cd265c7999f0b9a5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:42.016920Z","signature_b64":"FaidWf7NI6WmFvHiUcKi4Qoh65h0+bCYHQ8w9QF0baudELr1RgHm2i+73L1ffOSyc/4Ld5THJFyn5cpDk/hMBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f0730671e5482e3b7ad949d6212be45af54245bf836e839cf4cc83bf247bfd1","last_reissued_at":"2026-07-05T11:38:42.016408Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:42.016408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LoViC: Efficient Long Video Generation with Context Compression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Han Wu, Jiaxiu Jiang, Jingjing Ren, Wangmeng Zuo, Wenbo Li, Xiaogang Xu, Yong Guo, Yuping Qiu","submitted_at":"2025-07-17T09:46:43Z","abstract_excerpt":"Despite recent advances in diffusion transformers (DiTs) for text-to-video generation, scaling to long-duration content remains challenging due to the quadratic complexity of self-attention. While prior efforts -- such as sparse attention and temporally autoregressive models -- offer partial relief, they often compromise temporal coherence or scalability. We introduce LoViC, a DiT-based framework trained on million-scale open-domain videos, designed to produce long, coherent videos through a segment-wise generation process. At the core of our approach is FlexFormer, an expressive autoencoder t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.12952","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/2507.12952/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":"2507.12952","created_at":"2026-07-05T11:38:42.016469+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.12952v1","created_at":"2026-07-05T11:38:42.016469+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.12952","created_at":"2026-07-05T11:38:42.016469+00:00"},{"alias_kind":"pith_short_12","alias_value":"P4DTAZY6KSBO","created_at":"2026-07-05T11:38:42.016469+00:00"},{"alias_kind":"pith_short_16","alias_value":"P4DTAZY6KSBOHN5N","created_at":"2026-07-05T11:38:42.016469+00:00"},{"alias_kind":"pith_short_8","alias_value":"P4DTAZY6","created_at":"2026-07-05T11:38:42.016469+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04527","citing_title":"Echo-Infinity: Learning Evolving Memory for Real-Time Infinite Video Generation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23610","citing_title":"EM-Vid: Training-Free Entity-Centric Memory for Efficient and Consistent Multi-Shot Video Generation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18733","citing_title":"Advancing Narrative Long Video Generation via Training-Free Identity-Aware Memory","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2510.02283","citing_title":"Self-Forcing++: Towards Minute-Scale High-Quality Video Generation","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2603.28489","citing_title":"Video Generation Models as World Models: Efficient Paradigms, Architectures and Algorithms","ref_index":102,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15911","citing_title":"Efficient Video Diffusion Models: Advancements and Challenges","ref_index":60,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW","json":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW.json","graph_json":"https://pith.science/api/pith-number/P4DTAZY6KSBOHN5NSSOWEEV6IW/graph.json","events_json":"https://pith.science/api/pith-number/P4DTAZY6KSBOHN5NSSOWEEV6IW/events.json","paper":"https://pith.science/paper/P4DTAZY6"},"agent_actions":{"view_html":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW","download_json":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW.json","view_paper":"https://pith.science/paper/P4DTAZY6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.12952&json=true","fetch_graph":"https://pith.science/api/pith-number/P4DTAZY6KSBOHN5NSSOWEEV6IW/graph.json","fetch_events":"https://pith.science/api/pith-number/P4DTAZY6KSBOHN5NSSOWEEV6IW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW/action/storage_attestation","attest_author":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW/action/author_attestation","sign_citation":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW/action/citation_signature","submit_replication":"https://pith.science/pith/P4DTAZY6KSBOHN5NSSOWEEV6IW/action/replication_record"}},"created_at":"2026-07-05T11:38:42.016469+00:00","updated_at":"2026-07-05T11:38:42.016469+00:00"}