{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3Z4RIBAPUF65PXK3SU2HAJ73CT","short_pith_number":"pith:3Z4RIBAP","schema_version":"1.0","canonical_sha256":"de7914040fa17dd7dd5b95347027fb14e2947444dfc8a43a7c74a715e29eecf8","source":{"kind":"arxiv","id":"2602.20360","version":2},"attestation_state":"computed","paper":{"title":"Momentum Guidance: Plug-and-Play Guidance for Flow Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Baiyu Su, Chi Zhang, Jian Yu, Lizhang Chen, Qiang Liu, Runlong Liao","submitted_at":"2026-02-23T21:06:35Z","abstract_excerpt":"Flow-based generative methods offer a simple and effective framework for high-fidelity generation, yet pretrained flow models are rarely used in their vanilla conditional form: in image generation, samples without guidance often appear diffuse and lack fine-grained detail. Existing guidance techniques such as classifier-free guidance (CFG) improve fidelity but reduce sample diversity. We introduce Momentum Guidance (MG), a guidance method that improves sample quality by extrapolating the current velocity away from an exponential moving average of past velocities along the ODE trajectory, while"},"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":"2602.20360","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-23T21:06:35Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"584fa5c4d8af6f4375fe625a2081032e3ba393f8299f60d8d93c76e84172ec00","abstract_canon_sha256":"e29a6d3d39fe49823be13b05a1b23fd30e424be3312835822d16168f8ccd121b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:18:18.089238Z","signature_b64":"TCd6d77mKU2AdnvYbsKUQsUhA3KxLKXQwTuoFh49wnAhEzzBnTxDGoBy/qhIlluQWHuxtKQNsdn0IWfk7KcSDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de7914040fa17dd7dd5b95347027fb14e2947444dfc8a43a7c74a715e29eecf8","last_reissued_at":"2026-06-30T01:18:18.088518Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:18:18.088518Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Momentum Guidance: Plug-and-Play Guidance for Flow Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Baiyu Su, Chi Zhang, Jian Yu, Lizhang Chen, Qiang Liu, Runlong Liao","submitted_at":"2026-02-23T21:06:35Z","abstract_excerpt":"Flow-based generative methods offer a simple and effective framework for high-fidelity generation, yet pretrained flow models are rarely used in their vanilla conditional form: in image generation, samples without guidance often appear diffuse and lack fine-grained detail. Existing guidance techniques such as classifier-free guidance (CFG) improve fidelity but reduce sample diversity. We introduce Momentum Guidance (MG), a guidance method that improves sample quality by extrapolating the current velocity away from an exponential moving average of past velocities along the ODE trajectory, while"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.20360","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/2602.20360/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":"2602.20360","created_at":"2026-06-30T01:18:18.088603+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.20360v2","created_at":"2026-06-30T01:18:18.088603+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.20360","created_at":"2026-06-30T01:18:18.088603+00:00"},{"alias_kind":"pith_short_12","alias_value":"3Z4RIBAPUF65","created_at":"2026-06-30T01:18:18.088603+00:00"},{"alias_kind":"pith_short_16","alias_value":"3Z4RIBAPUF65PXK3","created_at":"2026-06-30T01:18:18.088603+00:00"},{"alias_kind":"pith_short_8","alias_value":"3Z4RIBAP","created_at":"2026-06-30T01:18:18.088603+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.23872","citing_title":"Training-Free Looped Transformers","ref_index":52,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT","json":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT.json","graph_json":"https://pith.science/api/pith-number/3Z4RIBAPUF65PXK3SU2HAJ73CT/graph.json","events_json":"https://pith.science/api/pith-number/3Z4RIBAPUF65PXK3SU2HAJ73CT/events.json","paper":"https://pith.science/paper/3Z4RIBAP"},"agent_actions":{"view_html":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT","download_json":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT.json","view_paper":"https://pith.science/paper/3Z4RIBAP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.20360&json=true","fetch_graph":"https://pith.science/api/pith-number/3Z4RIBAPUF65PXK3SU2HAJ73CT/graph.json","fetch_events":"https://pith.science/api/pith-number/3Z4RIBAPUF65PXK3SU2HAJ73CT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT/action/storage_attestation","attest_author":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT/action/author_attestation","sign_citation":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT/action/citation_signature","submit_replication":"https://pith.science/pith/3Z4RIBAPUF65PXK3SU2HAJ73CT/action/replication_record"}},"created_at":"2026-06-30T01:18:18.088603+00:00","updated_at":"2026-06-30T01:18:18.088603+00:00"}