{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EUQZ4CQ3G2AQXEN3IA5K3FIFK7","short_pith_number":"pith:EUQZ4CQ3","schema_version":"1.0","canonical_sha256":"25219e0a1b36810b91bb403aad950557e6383041161740b8dd9055faf5c91a21","source":{"kind":"arxiv","id":"2505.06603","version":1},"attestation_state":"computed","paper":{"title":"ReplayCAD: Generative Diffusion Replay for Continual Anomaly Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jinglin Liang, Lei Hu, Ling Deng, Lingyu Liang, Shuangping Huang, Tianshui Chen, Zhiyong Gan","submitted_at":"2025-05-10T11:29:03Z","abstract_excerpt":"Continual Anomaly Detection (CAD) enables anomaly detection models in learning new classes while preserving knowledge of historical classes. CAD faces two key challenges: catastrophic forgetting and segmentation of small anomalous regions. Existing CAD methods store image distributions or patch features to mitigate catastrophic forgetting, but they fail to preserve pixel-level detailed features for accurate segmentation. To overcome this limitation, we propose ReplayCAD, a novel diffusion-driven generative replay framework that replay high-quality historical data, thus effectively preserving p"},"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":"2505.06603","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-10T11:29:03Z","cross_cats_sorted":[],"title_canon_sha256":"8f3e2ed7199738c49132cc5d2f1c486b77cfa99fcc9c8883e9a06a873b2168fa","abstract_canon_sha256":"21ca1f5c1138d9355e5fbf5f9b01b4a1d07e9fba1a344cfc6db6fd2dc5387fa4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:01:12.922034Z","signature_b64":"IfOMoCIzgYJy96rhI/4IcCrQD91sPzEiNY1TsX3w6P36djWbHYuiTLEGtJYSXa+Wa9y1WhK9t1UBn0A4GUEDAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25219e0a1b36810b91bb403aad950557e6383041161740b8dd9055faf5c91a21","last_reissued_at":"2026-07-05T11:01:12.921469Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:01:12.921469Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ReplayCAD: Generative Diffusion Replay for Continual Anomaly Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jinglin Liang, Lei Hu, Ling Deng, Lingyu Liang, Shuangping Huang, Tianshui Chen, Zhiyong Gan","submitted_at":"2025-05-10T11:29:03Z","abstract_excerpt":"Continual Anomaly Detection (CAD) enables anomaly detection models in learning new classes while preserving knowledge of historical classes. CAD faces two key challenges: catastrophic forgetting and segmentation of small anomalous regions. Existing CAD methods store image distributions or patch features to mitigate catastrophic forgetting, but they fail to preserve pixel-level detailed features for accurate segmentation. To overcome this limitation, we propose ReplayCAD, a novel diffusion-driven generative replay framework that replay high-quality historical data, thus effectively preserving p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.06603","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/2505.06603/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":"2505.06603","created_at":"2026-07-05T11:01:12.921529+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.06603v1","created_at":"2026-07-05T11:01:12.921529+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.06603","created_at":"2026-07-05T11:01:12.921529+00:00"},{"alias_kind":"pith_short_12","alias_value":"EUQZ4CQ3G2AQ","created_at":"2026-07-05T11:01:12.921529+00:00"},{"alias_kind":"pith_short_16","alias_value":"EUQZ4CQ3G2AQXEN3","created_at":"2026-07-05T11:01:12.921529+00:00"},{"alias_kind":"pith_short_8","alias_value":"EUQZ4CQ3","created_at":"2026-07-05T11:01:12.921529+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26687","citing_title":"DeCoFlow: Structural Decomposition of Normalizing Flows for Continual Anomaly Detection","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02042","citing_title":"Normality-Preserving Continual Industrial Anomaly Detection via Orthogonal LoRA Banks","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7","json":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7.json","graph_json":"https://pith.science/api/pith-number/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/graph.json","events_json":"https://pith.science/api/pith-number/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/events.json","paper":"https://pith.science/paper/EUQZ4CQ3"},"agent_actions":{"view_html":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7","download_json":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7.json","view_paper":"https://pith.science/paper/EUQZ4CQ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.06603&json=true","fetch_graph":"https://pith.science/api/pith-number/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/graph.json","fetch_events":"https://pith.science/api/pith-number/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/action/storage_attestation","attest_author":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/action/author_attestation","sign_citation":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/action/citation_signature","submit_replication":"https://pith.science/pith/EUQZ4CQ3G2AQXEN3IA5K3FIFK7/action/replication_record"}},"created_at":"2026-07-05T11:01:12.921529+00:00","updated_at":"2026-07-05T11:01:12.921529+00:00"}