{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SGMDLTJOGWIBJ6CBMV3DCTTZO6","short_pith_number":"pith:SGMDLTJO","schema_version":"1.0","canonical_sha256":"919835cd2e359014f8416576314e7977981398054024de14f24d83fbcac9a713","source":{"kind":"arxiv","id":"2505.05829","version":1},"attestation_state":"computed","paper":{"title":"Accelerating Diffusion Transformer via Increment-Calibrated Caching with Channel-Aware Singular Value Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Keyi Li, Le Ye, Yifan Jia, Yufei Ma, Zhiyuan Chen","submitted_at":"2025-05-09T06:56:17Z","abstract_excerpt":"Diffusion transformer (DiT) models have achieved remarkable success in image generation, thanks for their exceptional generative capabilities and scalability. Nonetheless, the iterative nature of diffusion models (DMs) results in high computation complexity, posing challenges for deployment. Although existing cache-based acceleration methods try to utilize the inherent temporal similarity to skip redundant computations of DiT, the lack of correction may induce potential quality degradation. In this paper, we propose increment-calibrated caching, a training-free method for DiT acceleration, whe"},"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":"2505.05829","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-09T06:56:17Z","cross_cats_sorted":["cs.LG","eess.IV"],"title_canon_sha256":"2fe91f55a6f81361dfb7e5aabee29a66aba9f5bc2ab9376b1bf307125128b95f","abstract_canon_sha256":"5dfc6cbd8e51401ec01c6a8b57634a264bba8a927aa0a9339341fa3e0890ea1a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:48.390740Z","signature_b64":"iwyNmAucwQFourMmr/XWjcBZy1jFlZV6nM3af/pBAvd8hICZSluclfzQiX4iei0GjC5uG3ecN50fm4dOvqvzCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"919835cd2e359014f8416576314e7977981398054024de14f24d83fbcac9a713","last_reissued_at":"2026-07-05T11:00:48.390175Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:48.390175Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accelerating Diffusion Transformer via Increment-Calibrated Caching with Channel-Aware Singular Value Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Keyi Li, Le Ye, Yifan Jia, Yufei Ma, Zhiyuan Chen","submitted_at":"2025-05-09T06:56:17Z","abstract_excerpt":"Diffusion transformer (DiT) models have achieved remarkable success in image generation, thanks for their exceptional generative capabilities and scalability. Nonetheless, the iterative nature of diffusion models (DMs) results in high computation complexity, posing challenges for deployment. Although existing cache-based acceleration methods try to utilize the inherent temporal similarity to skip redundant computations of DiT, the lack of correction may induce potential quality degradation. In this paper, we propose increment-calibrated caching, a training-free method for DiT acceleration, whe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.05829","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/2505.05829/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":"2505.05829","created_at":"2026-07-05T11:00:48.390234+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.05829v1","created_at":"2026-07-05T11:00:48.390234+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.05829","created_at":"2026-07-05T11:00:48.390234+00:00"},{"alias_kind":"pith_short_12","alias_value":"SGMDLTJOGWIB","created_at":"2026-07-05T11:00:48.390234+00:00"},{"alias_kind":"pith_short_16","alias_value":"SGMDLTJOGWIBJ6CB","created_at":"2026-07-05T11:00:48.390234+00:00"},{"alias_kind":"pith_short_8","alias_value":"SGMDLTJO","created_at":"2026-07-05T11:00:48.390234+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.24870","citing_title":"Trajectory-Consistent Calibration for Cache-Accelerated Diffusion Models","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25798","citing_title":"DiSC: Resolution-Scalable Acceleration of Diffusion Models by Exploiting Sparsity and Cached Token Reuse with Hash-based Distribution","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6","json":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6.json","graph_json":"https://pith.science/api/pith-number/SGMDLTJOGWIBJ6CBMV3DCTTZO6/graph.json","events_json":"https://pith.science/api/pith-number/SGMDLTJOGWIBJ6CBMV3DCTTZO6/events.json","paper":"https://pith.science/paper/SGMDLTJO"},"agent_actions":{"view_html":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6","download_json":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6.json","view_paper":"https://pith.science/paper/SGMDLTJO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.05829&json=true","fetch_graph":"https://pith.science/api/pith-number/SGMDLTJOGWIBJ6CBMV3DCTTZO6/graph.json","fetch_events":"https://pith.science/api/pith-number/SGMDLTJOGWIBJ6CBMV3DCTTZO6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6/action/storage_attestation","attest_author":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6/action/author_attestation","sign_citation":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6/action/citation_signature","submit_replication":"https://pith.science/pith/SGMDLTJOGWIBJ6CBMV3DCTTZO6/action/replication_record"}},"created_at":"2026-07-05T11:00:48.390234+00:00","updated_at":"2026-07-05T11:00:48.390234+00:00"}