{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:U7QPUJUANIE6YPMG2AFR3SK274","short_pith_number":"pith:U7QPUJUA","schema_version":"1.0","canonical_sha256":"a7e0fa26806a09ec3d86d00b1dc95aff0417c6bcd7b25e78a9ab88d3bd4e378e","source":{"kind":"arxiv","id":"2012.06765","version":1},"attestation_state":"computed","paper":{"title":"Anomaly detection through latent space restoration using vector-quantized variational autoencoders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Giacomo Tarroni, Sergio Naval Marimont","submitted_at":"2020-12-12T09:19:59Z","abstract_excerpt":"We propose an out-of-distribution detection method that combines density and restoration-based approaches using Vector-Quantized Variational Auto-Encoders (VQ-VAEs). The VQ-VAE model learns to encode images in a categorical latent space. The prior distribution of latent codes is then modelled using an Auto-Regressive (AR) model. We found that the prior probability estimated by the AR model can be useful for unsupervised anomaly detection and enables the estimation of both sample and pixel-wise anomaly scores. The sample-wise score is defined as the negative log-likelihood of the latent variabl"},"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":"2012.06765","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2020-12-12T09:19:59Z","cross_cats_sorted":["cs.LG","eess.IV"],"title_canon_sha256":"5c69b2254733e2d9d07ffea081bee6d772b67611a476af2be68105dcfad1fac5","abstract_canon_sha256":"799dd1cc2053185575dec1dd82932affd95573ad8bddeb43c819060b5ae9b51f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:58:59.335146Z","signature_b64":"cjG8W1IxMI2UyQnaSfL/kto9oD6bIcGizaCCeIH4GNLgj+nDoV64zhu+N4dm8x2zDJxst2sfCtl+LD6HLVHaDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a7e0fa26806a09ec3d86d00b1dc95aff0417c6bcd7b25e78a9ab88d3bd4e378e","last_reissued_at":"2026-07-05T01:58:59.334690Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:58:59.334690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Anomaly detection through latent space restoration using vector-quantized variational autoencoders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Giacomo Tarroni, Sergio Naval Marimont","submitted_at":"2020-12-12T09:19:59Z","abstract_excerpt":"We propose an out-of-distribution detection method that combines density and restoration-based approaches using Vector-Quantized Variational Auto-Encoders (VQ-VAEs). The VQ-VAE model learns to encode images in a categorical latent space. The prior distribution of latent codes is then modelled using an Auto-Regressive (AR) model. We found that the prior probability estimated by the AR model can be useful for unsupervised anomaly detection and enables the estimation of both sample and pixel-wise anomaly scores. The sample-wise score is defined as the negative log-likelihood of the latent variabl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.06765","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/2012.06765/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":"2012.06765","created_at":"2026-07-05T01:58:59.334749+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.06765v1","created_at":"2026-07-05T01:58:59.334749+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.06765","created_at":"2026-07-05T01:58:59.334749+00:00"},{"alias_kind":"pith_short_12","alias_value":"U7QPUJUANIE6","created_at":"2026-07-05T01:58:59.334749+00:00"},{"alias_kind":"pith_short_16","alias_value":"U7QPUJUANIE6YPMG","created_at":"2026-07-05T01:58:59.334749+00:00"},{"alias_kind":"pith_short_8","alias_value":"U7QPUJUA","created_at":"2026-07-05T01:58:59.334749+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.08052","citing_title":"RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings for Weakly Supervised Anomaly Detection in Brain MRI Scans","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22139","citing_title":"Anatomy-Aware Unsupervised Detection and Localization of Retinal Abnormalities in Optical Coherence Tomography","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274","json":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274.json","graph_json":"https://pith.science/api/pith-number/U7QPUJUANIE6YPMG2AFR3SK274/graph.json","events_json":"https://pith.science/api/pith-number/U7QPUJUANIE6YPMG2AFR3SK274/events.json","paper":"https://pith.science/paper/U7QPUJUA"},"agent_actions":{"view_html":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274","download_json":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274.json","view_paper":"https://pith.science/paper/U7QPUJUA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.06765&json=true","fetch_graph":"https://pith.science/api/pith-number/U7QPUJUANIE6YPMG2AFR3SK274/graph.json","fetch_events":"https://pith.science/api/pith-number/U7QPUJUANIE6YPMG2AFR3SK274/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274/action/storage_attestation","attest_author":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274/action/author_attestation","sign_citation":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274/action/citation_signature","submit_replication":"https://pith.science/pith/U7QPUJUANIE6YPMG2AFR3SK274/action/replication_record"}},"created_at":"2026-07-05T01:58:59.334749+00:00","updated_at":"2026-07-05T01:58:59.334749+00:00"}