{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:T7B3O44OKS4ZDDNQI6JPTPE3Q2","short_pith_number":"pith:T7B3O44O","schema_version":"1.0","canonical_sha256":"9fc3b7738e54b9918db04792f9bc9b868ec0f35a005c83ff53f5b6a3f62029f1","source":{"kind":"arxiv","id":"2605.23070","version":1},"attestation_state":"computed","paper":{"title":"Flow Mismatching: Unsupervised Anomaly Detection via Velocity Discrepancies in Flow Matching Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hao Yan, Kamran Paynabar, Mehrdad Moradi, Shengzhe Chen","submitted_at":"2026-05-21T22:01:31Z","abstract_excerpt":"We propose Flow Mismatching, an unsupervised anomaly detection method that deliberately avoids reconstruction-based paradigms. Instead, we treat flow matching as geometric dynamics and leverage a key insight: anomalies occur at places where the learned normal flow disagrees with the geometric path toward a test image.\n  Given a flow matching model trained only on normal images, we probe its learned velocity field along affine paths from Gaussian noise to a target image. Along each path, we compare the model-predicted velocity, which follows normal generative dynamics, with the geometric veloci"},"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":"2605.23070","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-21T22:01:31Z","cross_cats_sorted":[],"title_canon_sha256":"8d4b42785ff2aff97670de4a22d3a16daeddb5b550c57ed117793499fe2f02b7","abstract_canon_sha256":"045f612790194d121fff21897f265309932c75fff9d58ad26ea0d7f3a7c05fe4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:37.131763Z","signature_b64":"aWwl+H+wZwuHNLVVq2OTY48GNEyQq1SGGPEBMMUJPEkTFVPsHOL6lLJCyVdqFGQglVlgb6A+7qnHNJdF7iUYBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9fc3b7738e54b9918db04792f9bc9b868ec0f35a005c83ff53f5b6a3f62029f1","last_reissued_at":"2026-05-25T02:01:37.131330Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:37.131330Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Flow Mismatching: Unsupervised Anomaly Detection via Velocity Discrepancies in Flow Matching Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hao Yan, Kamran Paynabar, Mehrdad Moradi, Shengzhe Chen","submitted_at":"2026-05-21T22:01:31Z","abstract_excerpt":"We propose Flow Mismatching, an unsupervised anomaly detection method that deliberately avoids reconstruction-based paradigms. Instead, we treat flow matching as geometric dynamics and leverage a key insight: anomalies occur at places where the learned normal flow disagrees with the geometric path toward a test image.\n  Given a flow matching model trained only on normal images, we probe its learned velocity field along affine paths from Gaussian noise to a target image. Along each path, we compare the model-predicted velocity, which follows normal generative dynamics, with the geometric veloci"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.23070","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/2605.23070/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":"2605.23070","created_at":"2026-05-25T02:01:37.131394+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.23070v1","created_at":"2026-05-25T02:01:37.131394+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.23070","created_at":"2026-05-25T02:01:37.131394+00:00"},{"alias_kind":"pith_short_12","alias_value":"T7B3O44OKS4Z","created_at":"2026-05-25T02:01:37.131394+00:00"},{"alias_kind":"pith_short_16","alias_value":"T7B3O44OKS4ZDDNQ","created_at":"2026-05-25T02:01:37.131394+00:00"},{"alias_kind":"pith_short_8","alias_value":"T7B3O44O","created_at":"2026-05-25T02:01:37.131394+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/T7B3O44OKS4ZDDNQI6JPTPE3Q2","json":"https://pith.science/pith/T7B3O44OKS4ZDDNQI6JPTPE3Q2.json","graph_json":"https://pith.science/api/pith-number/T7B3O44OKS4ZDDNQI6JPTPE3Q2/graph.json","events_json":"https://pith.science/api/pith-number/T7B3O44OKS4ZDDNQI6JPTPE3Q2/events.json","paper":"https://pith.science/paper/T7B3O44O"},"agent_actions":{"view_html":"https://pith.science/pith/T7B3O44OKS4ZDDNQI6JPTPE3Q2","download_json":"https://pith.science/pith/T7B3O44OKS4ZDDNQI6JPTPE3Q2.json","view_paper":"https://pith.science/paper/T7B3O44O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.23070&json=true","fetch_graph":"https://pith.science/api/pith-number/T7B3O44OKS4ZDDNQI6JPTPE3Q2/graph.json","fetch_events":"https://pith.science/api/pith-number/T7B3O44OKS4ZDDNQI6JPTPE3Q2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T7B3O44OKS4ZDDNQI6JPTPE3Q2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T7B3O44OKS4ZDDNQI6JPTPE3Q2/action/storage_attestation","attest_author":"https://pith.science/pith/T7B3O44OKS4ZDDNQI6JPTPE3Q2/action/author_attestation","sign_citation":"https://pith.science/pith/T7B3O44OKS4ZDDNQI6JPTPE3Q2/action/citation_signature","submit_replication":"https://pith.science/pith/T7B3O44OKS4ZDDNQI6JPTPE3Q2/action/replication_record"}},"created_at":"2026-05-25T02:01:37.131394+00:00","updated_at":"2026-05-25T02:01:37.131394+00:00"}