{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:Y6SBAW7WYP63SXAQPQOGEXIJQZ","short_pith_number":"pith:Y6SBAW7W","schema_version":"1.0","canonical_sha256":"c7a4105bf6c3fdb95c107c1c625d09864f5ee22ba27262c59b8da15073002426","source":{"kind":"arxiv","id":"2411.02397","version":2},"attestation_state":"computed","paper":{"title":"Adaptive Caching for Faster Video Generation with Diffusion Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chenyang Zhang, Ding Liu, Haozhe Liu, Kumara Kahatapitiya, Menglin Jia, Michael S. Ryoo, Sen He, Tian Xie","submitted_at":"2024-11-04T18:59:44Z","abstract_excerpt":"Generating temporally-consistent high-fidelity videos can be computationally expensive, especially over longer temporal spans. More-recent Diffusion Transformers (DiTs) -- despite making significant headway in this context -- have only heightened such challenges as they rely on larger models and heavier attention mechanisms, resulting in slower inference speeds. In this paper, we introduce a training-free method to accelerate video DiTs, termed Adaptive Caching (AdaCache), which is motivated by the fact that \"not all videos are created equal\": meaning, some videos require fewer denoising steps"},"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":"2411.02397","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-11-04T18:59:44Z","cross_cats_sorted":[],"title_canon_sha256":"af0e68a537bfc6ca361dc90d33b129776dd3d261fa8756b887c724b60f369c2a","abstract_canon_sha256":"788a3e07937ede9329e83695e882a7c75abdebc37f00fcc90eae73bd52a635df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:32:33.549790Z","signature_b64":"V6FOZOTR8oQ4U0ZQ6PXVlKTtUj/7OHY2O80N+aBDXHNeFbaQ3hH780gWTE9NLf+/9gZroSSXB+nlb+hJouBdCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c7a4105bf6c3fdb95c107c1c625d09864f5ee22ba27262c59b8da15073002426","last_reissued_at":"2026-07-05T09:32:33.549256Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:32:33.549256Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adaptive Caching for Faster Video Generation with Diffusion Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chenyang Zhang, Ding Liu, Haozhe Liu, Kumara Kahatapitiya, Menglin Jia, Michael S. Ryoo, Sen He, Tian Xie","submitted_at":"2024-11-04T18:59:44Z","abstract_excerpt":"Generating temporally-consistent high-fidelity videos can be computationally expensive, especially over longer temporal spans. More-recent Diffusion Transformers (DiTs) -- despite making significant headway in this context -- have only heightened such challenges as they rely on larger models and heavier attention mechanisms, resulting in slower inference speeds. In this paper, we introduce a training-free method to accelerate video DiTs, termed Adaptive Caching (AdaCache), which is motivated by the fact that \"not all videos are created equal\": meaning, some videos require fewer denoising steps"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.02397","kind":"arxiv","version":2},"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/2411.02397/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":"2411.02397","created_at":"2026-07-05T09:32:33.549325+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.02397v2","created_at":"2026-07-05T09:32:33.549325+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.02397","created_at":"2026-07-05T09:32:33.549325+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y6SBAW7WYP63","created_at":"2026-07-05T09:32:33.549325+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y6SBAW7WYP63SXAQ","created_at":"2026-07-05T09:32:33.549325+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y6SBAW7W","created_at":"2026-07-05T09:32:33.549325+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05981","citing_title":"Inverting the Streaming-Diffusion Bottleneck: Video-Rate MLLM-Conditioned Edit Diffusion on a Consumer GPU","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02586","citing_title":"Fewer, Better Frames: A Compute-Normalized Proof of Concept for Coherence-First World-Model Rendering with Model-Guided FSR4 Frame Generation","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24870","citing_title":"Trajectory-Consistent Calibration for Cache-Accelerated Diffusion Models","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22015","citing_title":"ORBIS: Output-Guided Token Reduction with Distribution-Aware Matching for Video Diffusion Acceleration","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21042","citing_title":"Dynamic Video Generation: Shaping Video Generation Across Time and Space","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18190","citing_title":"Dual-Rate Diffusion: Accelerating diffusion models with an interleaved heavy-light network","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2506.13456","citing_title":"Block-wise Adaptive Caching for Accelerating Diffusion Policy","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2508.17588","citing_title":"HERO: Hierarchical Extrapolation and Refresh for Efficient World Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05449","citing_title":"DisCa: Accelerating Video Diffusion Transformers with Distillation-Compatible Learnable Feature Caching","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2602.13357","citing_title":"AdaCorrection: Adaptive Offset Cache Correction for Accurate Diffusion Transformers","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2602.22575","citing_title":"S2O: Early Stopping for Sparse Attention via Online Permutation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10723","citing_title":"AllocMV: Optimal Resource Allocation for Music Video Generation via Structured Persistent State","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14561","citing_title":"CoCoDiff: Optimizing Collective Communications for Distributed Diffusion Transformer Inference Under Ulysses Sequence Parallelism","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15911","citing_title":"Efficient Video Diffusion Models: Advancements and Challenges","ref_index":63,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ","json":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ.json","graph_json":"https://pith.science/api/pith-number/Y6SBAW7WYP63SXAQPQOGEXIJQZ/graph.json","events_json":"https://pith.science/api/pith-number/Y6SBAW7WYP63SXAQPQOGEXIJQZ/events.json","paper":"https://pith.science/paper/Y6SBAW7W"},"agent_actions":{"view_html":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ","download_json":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ.json","view_paper":"https://pith.science/paper/Y6SBAW7W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.02397&json=true","fetch_graph":"https://pith.science/api/pith-number/Y6SBAW7WYP63SXAQPQOGEXIJQZ/graph.json","fetch_events":"https://pith.science/api/pith-number/Y6SBAW7WYP63SXAQPQOGEXIJQZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ/action/storage_attestation","attest_author":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ/action/author_attestation","sign_citation":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ/action/citation_signature","submit_replication":"https://pith.science/pith/Y6SBAW7WYP63SXAQPQOGEXIJQZ/action/replication_record"}},"created_at":"2026-07-05T09:32:33.549325+00:00","updated_at":"2026-07-05T09:32:33.549325+00:00"}