{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2FDM4FGZX6LKN5P5XFSTUXVCML","short_pith_number":"pith:2FDM4FGZ","schema_version":"1.0","canonical_sha256":"d146ce14d9bf96a6f5fdb9653a5ea262fed7bb35ebee1f029c5c140e2a158822","source":{"kind":"arxiv","id":"2502.07508","version":3},"attestation_state":"computed","paper":{"title":"Enhance-A-Video: Better Generated Video for Free","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Kaipeng Zhang, Kai Wang, Mengzhao Chen, Wenqi Shao, Xuanlei Zhao, Yang Luo, Yang You, Zhangyang Wang","submitted_at":"2025-02-11T12:22:35Z","abstract_excerpt":"DiT-based video generation has achieved remarkable results, but research into enhancing existing models remains relatively unexplored. In this work, we introduce a training-free approach to enhance the coherence and quality of DiT-based generated videos, named Enhance-A-Video. The core idea is enhancing the cross-frame correlations based on non-diagonal temporal attention distributions. Thanks to its simple design, our approach can be easily applied to most DiT-based video generation frameworks without any retraining or fine-tuning. Across various DiT-based video generation models, our approac"},"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":"2502.07508","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-11T12:22:35Z","cross_cats_sorted":[],"title_canon_sha256":"6919d1e30f6a2f9aa769e243a07dcb5cf3e0c5e02595743d4c8ba27c44a283bd","abstract_canon_sha256":"e1c6fcda67150219a57d06ca38f712e4cfa6d2a75b12f248ed17bc119bf76397"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:20:43.646803Z","signature_b64":"0Uik8kRdNH0UxsG1S5EbNW0DbkLxU130XhSWik8gi2AMczU4rOgCJh/hFIEH9Db/MUicn2L4HWnV5tN6zY5VDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d146ce14d9bf96a6f5fdb9653a5ea262fed7bb35ebee1f029c5c140e2a158822","last_reissued_at":"2026-07-05T10:20:43.645032Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:20:43.645032Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhance-A-Video: Better Generated Video for Free","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Kaipeng Zhang, Kai Wang, Mengzhao Chen, Wenqi Shao, Xuanlei Zhao, Yang Luo, Yang You, Zhangyang Wang","submitted_at":"2025-02-11T12:22:35Z","abstract_excerpt":"DiT-based video generation has achieved remarkable results, but research into enhancing existing models remains relatively unexplored. In this work, we introduce a training-free approach to enhance the coherence and quality of DiT-based generated videos, named Enhance-A-Video. The core idea is enhancing the cross-frame correlations based on non-diagonal temporal attention distributions. Thanks to its simple design, our approach can be easily applied to most DiT-based video generation frameworks without any retraining or fine-tuning. Across various DiT-based video generation models, our approac"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.07508","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/2502.07508/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":"2502.07508","created_at":"2026-07-05T10:20:43.646222+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.07508v3","created_at":"2026-07-05T10:20:43.646222+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.07508","created_at":"2026-07-05T10:20:43.646222+00:00"},{"alias_kind":"pith_short_12","alias_value":"2FDM4FGZX6LK","created_at":"2026-07-05T10:20:43.646222+00:00"},{"alias_kind":"pith_short_16","alias_value":"2FDM4FGZX6LKN5P5","created_at":"2026-07-05T10:20:43.646222+00:00"},{"alias_kind":"pith_short_8","alias_value":"2FDM4FGZ","created_at":"2026-07-05T10:20:43.646222+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11969","citing_title":"SpecLoR: Spectral Lookahead Rectification for Motion-Coherent Text-to-Video Generation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2504.17180","citing_title":"We'll Fix it in Post: Improving Text-to-Video Generation with Neuro-Symbolic Feedback","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16503","citing_title":"Motif-Video 2B: Technical Report","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":41,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16503","citing_title":"Motif-Video 2B: Technical Report","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14910","citing_title":"Reward-Aware Trajectory Shaping for Few-step Visual Generation","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML","json":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML.json","graph_json":"https://pith.science/api/pith-number/2FDM4FGZX6LKN5P5XFSTUXVCML/graph.json","events_json":"https://pith.science/api/pith-number/2FDM4FGZX6LKN5P5XFSTUXVCML/events.json","paper":"https://pith.science/paper/2FDM4FGZ"},"agent_actions":{"view_html":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML","download_json":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML.json","view_paper":"https://pith.science/paper/2FDM4FGZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.07508&json=true","fetch_graph":"https://pith.science/api/pith-number/2FDM4FGZX6LKN5P5XFSTUXVCML/graph.json","fetch_events":"https://pith.science/api/pith-number/2FDM4FGZX6LKN5P5XFSTUXVCML/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML/action/storage_attestation","attest_author":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML/action/author_attestation","sign_citation":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML/action/citation_signature","submit_replication":"https://pith.science/pith/2FDM4FGZX6LKN5P5XFSTUXVCML/action/replication_record"}},"created_at":"2026-07-05T10:20:43.646222+00:00","updated_at":"2026-07-05T10:20:43.646222+00:00"}