{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BPZB3OHIKGIYXMJGHX7QFBAFMV","short_pith_number":"pith:BPZB3OHI","schema_version":"1.0","canonical_sha256":"0bf21db8e851918bb1263dff0284056554661e20552fa154f622207d8516802a","source":{"kind":"arxiv","id":"2601.21892","version":2},"attestation_state":"computed","paper":{"title":"Improving Classifier-Free Guidance of Flow Matching via Manifold Projection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Reformulating classifier-free guidance in flow matching as manifold-constrained homotopy optimization reduces sensitivity to guidance scales.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chao Wang, Haixia Liu, Jian-Feng Cai, Zhengyi Su","submitted_at":"2026-01-29T15:49:31Z","abstract_excerpt":"Classifier-free guidance (CFG) is a widely used technique for controllable generation in diffusion and flow-based models. Despite its empirical success, CFG relies on a heuristic linear extrapolation that is often sensitive to the guidance scale. In this work, we provide a principled interpretation of CFG through the lens of optimization. We demonstrate that the velocity field in flow matching corresponds to the gradient of a sequence of smoothed distance functions, which guides latent variables toward the scaled target image set. This perspective reveals that the standard CFG formulation is a"},"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":true},"canonical_record":{"source":{"id":"2601.21892","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-01-29T15:49:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c45fbbc391060580efa72447877bc0a05d7db1f3dd9dfd57f8b10877e43ac96c","abstract_canon_sha256":"5dbdb68d2c990ee0c961abb609095cfbceee3d44e1890a34583508fb81552a57"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:24.099959Z","signature_b64":"7Sa5o1a9puJAsbI653LKMH5JohmT8oU0tjhlfHLn8FvXzGqv4dbbh+Mn2wZnfBk0azeYg3MxTBidqgtJhb4+BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0bf21db8e851918bb1263dff0284056554661e20552fa154f622207d8516802a","last_reissued_at":"2026-05-18T03:09:24.099324Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:24.099324Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Classifier-Free Guidance of Flow Matching via Manifold Projection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Reformulating classifier-free guidance in flow matching as manifold-constrained homotopy optimization reduces sensitivity to guidance scales.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chao Wang, Haixia Liu, Jian-Feng Cai, Zhengyi Su","submitted_at":"2026-01-29T15:49:31Z","abstract_excerpt":"Classifier-free guidance (CFG) is a widely used technique for controllable generation in diffusion and flow-based models. Despite its empirical success, CFG relies on a heuristic linear extrapolation that is often sensitive to the guidance scale. In this work, we provide a principled interpretation of CFG through the lens of optimization. We demonstrate that the velocity field in flow matching corresponds to the gradient of a sequence of smoothed distance functions, which guides latent variables toward the scaled target image set. This perspective reveals that the standard CFG formulation is a"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We reformulate the CFG sampling as a homotopy optimization with a manifold constraint. This formulation necessitates a manifold projection step, which we implement via an incremental gradient descent scheme during sampling.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the velocity field in flow matching exactly corresponds to the gradient of a sequence of smoothed distance functions guiding latent variables toward the scaled target image set, making standard CFG merely an approximation whose gap controls sensitivity.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Reformulates CFG sampling in flow matching as homotopy optimization with manifold projection via incremental gradient descent and Anderson acceleration, yielding better fidelity and robustness without retraining.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Reformulating classifier-free guidance in flow matching as manifold-constrained homotopy optimization reduces sensitivity to guidance scales.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"218ada2be6aab54ce68536ae4ac14a9b30af0864b9230229b25bb89e9c662cde"},"source":{"id":"2601.21892","kind":"arxiv","version":2},"verdict":{"id":"2ec256b3-fd97-4606-9ddb-05647fedb311","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T09:40:42.595531Z","strongest_claim":"We reformulate the CFG sampling as a homotopy optimization with a manifold constraint. This formulation necessitates a manifold projection step, which we implement via an incremental gradient descent scheme during sampling.","one_line_summary":"Reformulates CFG sampling in flow matching as homotopy optimization with manifold projection via incremental gradient descent and Anderson acceleration, yielding better fidelity and robustness without retraining.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the velocity field in flow matching exactly corresponds to the gradient of a sequence of smoothed distance functions guiding latent variables toward the scaled target image set, making standard CFG merely an approximation whose gap controls sensitivity.","pith_extraction_headline":"Reformulating classifier-free guidance in flow matching as manifold-constrained homotopy optimization reduces sensitivity to guidance scales."},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"b7b7a2ff27ae5244c865e08bb4baf6e4acb4c71758fd46b10c733d768075f22b"},"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":"2601.21892","created_at":"2026-05-18T03:09:24.099434+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.21892v2","created_at":"2026-05-18T03:09:24.099434+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.21892","created_at":"2026-05-18T03:09:24.099434+00:00"},{"alias_kind":"pith_short_12","alias_value":"BPZB3OHIKGIY","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"BPZB3OHIKGIYXMJG","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"BPZB3OHI","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV","json":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV.json","graph_json":"https://pith.science/api/pith-number/BPZB3OHIKGIYXMJGHX7QFBAFMV/graph.json","events_json":"https://pith.science/api/pith-number/BPZB3OHIKGIYXMJGHX7QFBAFMV/events.json","paper":"https://pith.science/paper/BPZB3OHI"},"agent_actions":{"view_html":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV","download_json":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV.json","view_paper":"https://pith.science/paper/BPZB3OHI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.21892&json=true","fetch_graph":"https://pith.science/api/pith-number/BPZB3OHIKGIYXMJGHX7QFBAFMV/graph.json","fetch_events":"https://pith.science/api/pith-number/BPZB3OHIKGIYXMJGHX7QFBAFMV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV/action/storage_attestation","attest_author":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV/action/author_attestation","sign_citation":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV/action/citation_signature","submit_replication":"https://pith.science/pith/BPZB3OHIKGIYXMJGHX7QFBAFMV/action/replication_record"}},"created_at":"2026-05-18T03:09:24.099434+00:00","updated_at":"2026-05-18T03:09:24.099434+00:00"}