{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DHN24SMD3ULA63DRAMOO2YQLQS","short_pith_number":"pith:DHN24SMD","schema_version":"1.0","canonical_sha256":"19dbae4983dd160f6c71031ced620b8499fa5699d7cd8f9d7608035c9e76ac4b","source":{"kind":"arxiv","id":"2605.17244","version":1},"attestation_state":"computed","paper":{"title":"Drift Flow Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chenrui Ma, Ferdinando Fioretto, Lin Zhao, Tianyang Wang, Xi Xiao, Yanning Shen","submitted_at":"2026-05-17T03:58:01Z","abstract_excerpt":"Iterative generative models such as Flow Matching and Diffusion models have demonstrated strong test-time scaling behavior, where additional inference computation can improve generation quality. In contrast, Drift Models offer efficient one-step generation, but their direct generation paradigm limits such flexibility. In this work, we propose Drift Flow Matching (DFM), a framework that connects drifting generative modeling with flow-based iterative generation. DFM preserves the efficiency of direct transport maps while enabling generation to be refined through multiple inference steps when des"},"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":"2605.17244","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-17T03:58:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0611fbb3a1c9a4171f825f19973637ff976d143e6d593584d5227365a6e5fe60","abstract_canon_sha256":"b97b9407f8922c7b5a1a4ee419ce93c588e20aaee76aa274a3ebd55b774ec697"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:47.274398Z","signature_b64":"zbwcqTrckltDt+3YnlxQQjPro7bNSQu/7Tw6bgF0rV/drPyGtJ88UnVizStz5E7WdsSpMa5bPJptY/NN5YanDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"19dbae4983dd160f6c71031ced620b8499fa5699d7cd8f9d7608035c9e76ac4b","last_reissued_at":"2026-05-20T00:03:47.273478Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:47.273478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Drift Flow Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chenrui Ma, Ferdinando Fioretto, Lin Zhao, Tianyang Wang, Xi Xiao, Yanning Shen","submitted_at":"2026-05-17T03:58:01Z","abstract_excerpt":"Iterative generative models such as Flow Matching and Diffusion models have demonstrated strong test-time scaling behavior, where additional inference computation can improve generation quality. In contrast, Drift Models offer efficient one-step generation, but their direct generation paradigm limits such flexibility. In this work, we propose Drift Flow Matching (DFM), a framework that connects drifting generative modeling with flow-based iterative generation. DFM preserves the efficiency of direct transport maps while enabling generation to be refined through multiple inference steps when des"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17244","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/2605.17244/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-19T22:01:57.878310Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T21:33:23.794207Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"49aea23d476b1ca92a12c4f04716a266b203bd8e98d9851d285d915d9d4420a2"},"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":"2605.17244","created_at":"2026-05-20T00:03:47.273622+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.17244v1","created_at":"2026-05-20T00:03:47.273622+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17244","created_at":"2026-05-20T00:03:47.273622+00:00"},{"alias_kind":"pith_short_12","alias_value":"DHN24SMD3ULA","created_at":"2026-05-20T00:03:47.273622+00:00"},{"alias_kind":"pith_short_16","alias_value":"DHN24SMD3ULA63DR","created_at":"2026-05-20T00:03:47.273622+00:00"},{"alias_kind":"pith_short_8","alias_value":"DHN24SMD","created_at":"2026-05-20T00:03:47.273622+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/DHN24SMD3ULA63DRAMOO2YQLQS","json":"https://pith.science/pith/DHN24SMD3ULA63DRAMOO2YQLQS.json","graph_json":"https://pith.science/api/pith-number/DHN24SMD3ULA63DRAMOO2YQLQS/graph.json","events_json":"https://pith.science/api/pith-number/DHN24SMD3ULA63DRAMOO2YQLQS/events.json","paper":"https://pith.science/paper/DHN24SMD"},"agent_actions":{"view_html":"https://pith.science/pith/DHN24SMD3ULA63DRAMOO2YQLQS","download_json":"https://pith.science/pith/DHN24SMD3ULA63DRAMOO2YQLQS.json","view_paper":"https://pith.science/paper/DHN24SMD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.17244&json=true","fetch_graph":"https://pith.science/api/pith-number/DHN24SMD3ULA63DRAMOO2YQLQS/graph.json","fetch_events":"https://pith.science/api/pith-number/DHN24SMD3ULA63DRAMOO2YQLQS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DHN24SMD3ULA63DRAMOO2YQLQS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DHN24SMD3ULA63DRAMOO2YQLQS/action/storage_attestation","attest_author":"https://pith.science/pith/DHN24SMD3ULA63DRAMOO2YQLQS/action/author_attestation","sign_citation":"https://pith.science/pith/DHN24SMD3ULA63DRAMOO2YQLQS/action/citation_signature","submit_replication":"https://pith.science/pith/DHN24SMD3ULA63DRAMOO2YQLQS/action/replication_record"}},"created_at":"2026-05-20T00:03:47.273622+00:00","updated_at":"2026-05-20T00:03:47.273622+00:00"}