{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MSDNXOKN6MEUSAHYATSTN6JVPB","short_pith_number":"pith:MSDNXOKN","schema_version":"1.0","canonical_sha256":"6486dbb94df3094900f804e536f9357878c308718587015abf266154b36679ee","source":{"kind":"arxiv","id":"2507.14797","version":1},"attestation_state":"computed","paper":{"title":"Distilling Parallel Gradients for Fast ODE Solvers of Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Beier Zhu, Chi Zhang, Hanwang Zhang, Ruoyu Wang, Tong Zhao","submitted_at":"2025-07-20T03:08:06Z","abstract_excerpt":"Diffusion models (DMs) have achieved state-of-the-art generative performance but suffer from high sampling latency due to their sequential denoising nature. Existing solver-based acceleration methods often face image quality degradation under a low-latency budget. In this paper, we propose the Ensemble Parallel Direction solver (dubbed as \\ours), a novel ODE solver that mitigates truncation errors by incorporating multiple parallel gradient evaluations in each ODE step. Importantly, since the additional gradient computations are independent, they can be fully parallelized, preserving low-laten"},"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":"2507.14797","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-20T03:08:06Z","cross_cats_sorted":[],"title_canon_sha256":"72d3d38da2039811922ac53f412248d36e926e138e3a08642b67a4adf79ceba7","abstract_canon_sha256":"2c22615e1207fef0e8408e6f510ace4ca95e8843217e760b32d9d6cd26c74d7f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:40:13.715634Z","signature_b64":"A7iKuAostMcq/hWczT/LsYbf2iCyEtaLZmjnkybG5XLW0Xamwwsv8kgrjSuWOU/H/sNoXhGiO18iut00FeuQBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6486dbb94df3094900f804e536f9357878c308718587015abf266154b36679ee","last_reissued_at":"2026-07-05T11:40:13.715029Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:40:13.715029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distilling Parallel Gradients for Fast ODE Solvers of Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Beier Zhu, Chi Zhang, Hanwang Zhang, Ruoyu Wang, Tong Zhao","submitted_at":"2025-07-20T03:08:06Z","abstract_excerpt":"Diffusion models (DMs) have achieved state-of-the-art generative performance but suffer from high sampling latency due to their sequential denoising nature. Existing solver-based acceleration methods often face image quality degradation under a low-latency budget. In this paper, we propose the Ensemble Parallel Direction solver (dubbed as \\ours), a novel ODE solver that mitigates truncation errors by incorporating multiple parallel gradient evaluations in each ODE step. Importantly, since the additional gradient computations are independent, they can be fully parallelized, preserving low-laten"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.14797","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/2507.14797/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":"2507.14797","created_at":"2026-07-05T11:40:13.715096+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.14797v1","created_at":"2026-07-05T11:40:13.715096+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.14797","created_at":"2026-07-05T11:40:13.715096+00:00"},{"alias_kind":"pith_short_12","alias_value":"MSDNXOKN6MEU","created_at":"2026-07-05T11:40:13.715096+00:00"},{"alias_kind":"pith_short_16","alias_value":"MSDNXOKN6MEUSAHY","created_at":"2026-07-05T11:40:13.715096+00:00"},{"alias_kind":"pith_short_8","alias_value":"MSDNXOKN","created_at":"2026-07-05T11:40:13.715096+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB","json":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB.json","graph_json":"https://pith.science/api/pith-number/MSDNXOKN6MEUSAHYATSTN6JVPB/graph.json","events_json":"https://pith.science/api/pith-number/MSDNXOKN6MEUSAHYATSTN6JVPB/events.json","paper":"https://pith.science/paper/MSDNXOKN"},"agent_actions":{"view_html":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB","download_json":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB.json","view_paper":"https://pith.science/paper/MSDNXOKN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.14797&json=true","fetch_graph":"https://pith.science/api/pith-number/MSDNXOKN6MEUSAHYATSTN6JVPB/graph.json","fetch_events":"https://pith.science/api/pith-number/MSDNXOKN6MEUSAHYATSTN6JVPB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB/action/storage_attestation","attest_author":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB/action/author_attestation","sign_citation":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB/action/citation_signature","submit_replication":"https://pith.science/pith/MSDNXOKN6MEUSAHYATSTN6JVPB/action/replication_record"}},"created_at":"2026-07-05T11:40:13.715096+00:00","updated_at":"2026-07-05T11:40:13.715096+00:00"}