{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JGQBQVTTPFT7LJ5G75TEHGOENK","short_pith_number":"pith:JGQBQVTT","schema_version":"1.0","canonical_sha256":"49a01856737967f5a7a6ff664399c46abf563a0d44dadaa435c3369c2143f3e5","source":{"kind":"arxiv","id":"2503.14151","version":3},"attestation_state":"computed","paper":{"title":"Concat-ID: Towards Universal Identity-Preserving Video Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chongxuan Li, Jiayan Teng, Xiaotao Gu, Yong Zhong, Zhuoyi Yang","submitted_at":"2025-03-18T11:17:32Z","abstract_excerpt":"We present Concat-ID, a unified framework for identity-preserving video generation. Concat-ID employs variational autoencoders to extract image features, which are then concatenated with video latents along the sequence dimension. It relies exclusively on inherent 3D self-attention mechanisms to incorporate them, eliminating the need for additional parameters or modules. A novel cross-video pairing strategy and a multi-stage training regimen are introduced to balance identity consistency and facial editability while enhancing video naturalness. Extensive experiments demonstrate Concat-ID's sup"},"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":"2503.14151","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-03-18T11:17:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3ad6b8a3416a8d487287df49aa2354fa9efe2ef7d72bf888bec749af6e8384dd","abstract_canon_sha256":"2d10d40a41a7d4d84c324cef75f559a674b7d4aa247330792ea218fa06671ed4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:30:42.409747Z","signature_b64":"SuXdJpGifyf/Kj0VIvEGl+GTkjuyImvw50T/B9SlIkKTm7ZBIfe7vvaLCHKmmc+KzIBIzbo0v4C/ENkGYLJbAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"49a01856737967f5a7a6ff664399c46abf563a0d44dadaa435c3369c2143f3e5","last_reissued_at":"2026-07-05T11:30:42.409171Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:30:42.409171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Concat-ID: Towards Universal Identity-Preserving Video Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chongxuan Li, Jiayan Teng, Xiaotao Gu, Yong Zhong, Zhuoyi Yang","submitted_at":"2025-03-18T11:17:32Z","abstract_excerpt":"We present Concat-ID, a unified framework for identity-preserving video generation. Concat-ID employs variational autoencoders to extract image features, which are then concatenated with video latents along the sequence dimension. It relies exclusively on inherent 3D self-attention mechanisms to incorporate them, eliminating the need for additional parameters or modules. A novel cross-video pairing strategy and a multi-stage training regimen are introduced to balance identity consistency and facial editability while enhancing video naturalness. Extensive experiments demonstrate Concat-ID's sup"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.14151","kind":"arxiv","version":3},"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/2503.14151/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":"2503.14151","created_at":"2026-07-05T11:30:42.409237+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.14151v3","created_at":"2026-07-05T11:30:42.409237+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.14151","created_at":"2026-07-05T11:30:42.409237+00:00"},{"alias_kind":"pith_short_12","alias_value":"JGQBQVTTPFT7","created_at":"2026-07-05T11:30:42.409237+00:00"},{"alias_kind":"pith_short_16","alias_value":"JGQBQVTTPFT7LJ5G","created_at":"2026-07-05T11:30:42.409237+00:00"},{"alias_kind":"pith_short_8","alias_value":"JGQBQVTT","created_at":"2026-07-05T11:30:42.409237+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.04702","citing_title":"FaithfulFaces: Pose-Faithful Facial Identity Preservation for Text-to-Video Generation","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10030","citing_title":"Prompt Relay: Inference-Time Temporal Control for Multi-Event Video Generation","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK","json":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK.json","graph_json":"https://pith.science/api/pith-number/JGQBQVTTPFT7LJ5G75TEHGOENK/graph.json","events_json":"https://pith.science/api/pith-number/JGQBQVTTPFT7LJ5G75TEHGOENK/events.json","paper":"https://pith.science/paper/JGQBQVTT"},"agent_actions":{"view_html":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK","download_json":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK.json","view_paper":"https://pith.science/paper/JGQBQVTT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.14151&json=true","fetch_graph":"https://pith.science/api/pith-number/JGQBQVTTPFT7LJ5G75TEHGOENK/graph.json","fetch_events":"https://pith.science/api/pith-number/JGQBQVTTPFT7LJ5G75TEHGOENK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK/action/storage_attestation","attest_author":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK/action/author_attestation","sign_citation":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK/action/citation_signature","submit_replication":"https://pith.science/pith/JGQBQVTTPFT7LJ5G75TEHGOENK/action/replication_record"}},"created_at":"2026-07-05T11:30:42.409237+00:00","updated_at":"2026-07-05T11:30:42.409237+00:00"}