{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MZ53AL3TOK5OYJNFBBZMCKQLUI","short_pith_number":"pith:MZ53AL3T","schema_version":"1.0","canonical_sha256":"667bb02f7372baec25a50872c12a0ba2273cc70b007c6a09bb4da039762c972a","source":{"kind":"arxiv","id":"2408.08822","version":3},"attestation_state":"computed","paper":{"title":"PFDiff: Training-Free Acceleration of Diffusion Models Combining Past and Future Scores","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guangyi Wang, Lijiang Li, Songzhi Su, Wei Peng, Yuren Cai","submitted_at":"2024-08-16T16:12:44Z","abstract_excerpt":"Diffusion Probabilistic Models (DPMs) have shown remarkable potential in image generation, but their sampling efficiency is hindered by the need for numerous denoising steps. Most existing solutions accelerate the sampling process by proposing fast ODE solvers. However, the inevitable discretization errors of the ODE solvers are significantly magnified when the number of function evaluations (NFE) is fewer. In this work, we propose PFDiff, a novel training-free and orthogonal timestep-skipping strategy, which enables existing fast ODE solvers to operate with fewer NFE. Specifically, PFDiff ini"},"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":"2408.08822","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-08-16T16:12:44Z","cross_cats_sorted":[],"title_canon_sha256":"0810274dc9ccea13066db5b8383163e7e9c2e62cc04f29f4232de7a187489b43","abstract_canon_sha256":"4f429e7d2f2c8f841a3004b5cbbf9bb93697c758e16ec67b72d18f99cdb3cc86"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:17:10.790668Z","signature_b64":"OI1Ss0XMv3vlR20lTPxiQzrc2foJElNAZUiBZ9L2Y4LaepuB0QHGydbga66O8WVNegB1dQ31uqG92TNR1bidAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"667bb02f7372baec25a50872c12a0ba2273cc70b007c6a09bb4da039762c972a","last_reissued_at":"2026-07-05T10:17:10.790189Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:17:10.790189Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PFDiff: Training-Free Acceleration of Diffusion Models Combining Past and Future Scores","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guangyi Wang, Lijiang Li, Songzhi Su, Wei Peng, Yuren Cai","submitted_at":"2024-08-16T16:12:44Z","abstract_excerpt":"Diffusion Probabilistic Models (DPMs) have shown remarkable potential in image generation, but their sampling efficiency is hindered by the need for numerous denoising steps. Most existing solutions accelerate the sampling process by proposing fast ODE solvers. However, the inevitable discretization errors of the ODE solvers are significantly magnified when the number of function evaluations (NFE) is fewer. In this work, we propose PFDiff, a novel training-free and orthogonal timestep-skipping strategy, which enables existing fast ODE solvers to operate with fewer NFE. Specifically, PFDiff ini"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.08822","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/2408.08822/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":"2408.08822","created_at":"2026-07-05T10:17:10.790244+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.08822v3","created_at":"2026-07-05T10:17:10.790244+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.08822","created_at":"2026-07-05T10:17:10.790244+00:00"},{"alias_kind":"pith_short_12","alias_value":"MZ53AL3TOK5O","created_at":"2026-07-05T10:17:10.790244+00:00"},{"alias_kind":"pith_short_16","alias_value":"MZ53AL3TOK5OYJNF","created_at":"2026-07-05T10:17:10.790244+00:00"},{"alias_kind":"pith_short_8","alias_value":"MZ53AL3T","created_at":"2026-07-05T10:17:10.790244+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26769","citing_title":"ResilPhase: Plug-and-Play Phase Mapping and Noise-Resilient Macro-Trajectory Extrapolation for Diffusion Acceleration","ref_index":99,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI","json":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI.json","graph_json":"https://pith.science/api/pith-number/MZ53AL3TOK5OYJNFBBZMCKQLUI/graph.json","events_json":"https://pith.science/api/pith-number/MZ53AL3TOK5OYJNFBBZMCKQLUI/events.json","paper":"https://pith.science/paper/MZ53AL3T"},"agent_actions":{"view_html":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI","download_json":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI.json","view_paper":"https://pith.science/paper/MZ53AL3T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.08822&json=true","fetch_graph":"https://pith.science/api/pith-number/MZ53AL3TOK5OYJNFBBZMCKQLUI/graph.json","fetch_events":"https://pith.science/api/pith-number/MZ53AL3TOK5OYJNFBBZMCKQLUI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI/action/storage_attestation","attest_author":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI/action/author_attestation","sign_citation":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI/action/citation_signature","submit_replication":"https://pith.science/pith/MZ53AL3TOK5OYJNFBBZMCKQLUI/action/replication_record"}},"created_at":"2026-07-05T10:17:10.790244+00:00","updated_at":"2026-07-05T10:17:10.790244+00:00"}