{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:NJGJI6B3UMRUZRX4FUJCFXD6GF","short_pith_number":"pith:NJGJI6B3","schema_version":"1.0","canonical_sha256":"6a4c94783ba3234cc6fc2d1222dc7e314b6f24ef24813ec31a7e6c869decbcd2","source":{"kind":"arxiv","id":"2010.13064","version":1},"attestation_state":"computed","paper":{"title":"Further Analysis of Outlier Detection with Deep Generative Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Bin Dai, David Wipf, Jun Zhu, Ziyu Wang","submitted_at":"2020-10-25T08:20:38Z","abstract_excerpt":"The recent, counter-intuitive discovery that deep generative models (DGMs) can frequently assign a higher likelihood to outliers has implications for both outlier detection applications as well as our overall understanding of generative modeling. In this work, we present a possible explanation for this phenomenon, starting from the observation that a model's typical set and high-density region may not conincide. From this vantage point we propose a novel outlier test, the empirical success of which suggests that the failure of existing likelihood-based outlier tests does not necessarily imply "},"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":"2010.13064","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-10-25T08:20:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3ce62744488a4af2d1bd22d8b4e759aee0861b9aa59e3154bb15e3de8ea2c575","abstract_canon_sha256":"d6a49c89afef24b5b65f62e84a160bc035be64ff8e42aff85680bf918497410d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:45:51.685586Z","signature_b64":"NuSho7BQjF+lw0ws8LjIag0wEwyqIaKPIDEeIhndl2sP0549pznXecJnbQk3lTGO66LOJRfcgjTUrSxIdx19Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a4c94783ba3234cc6fc2d1222dc7e314b6f24ef24813ec31a7e6c869decbcd2","last_reissued_at":"2026-07-05T01:45:51.685226Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:45:51.685226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Further Analysis of Outlier Detection with Deep Generative Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Bin Dai, David Wipf, Jun Zhu, Ziyu Wang","submitted_at":"2020-10-25T08:20:38Z","abstract_excerpt":"The recent, counter-intuitive discovery that deep generative models (DGMs) can frequently assign a higher likelihood to outliers has implications for both outlier detection applications as well as our overall understanding of generative modeling. In this work, we present a possible explanation for this phenomenon, starting from the observation that a model's typical set and high-density region may not conincide. From this vantage point we propose a novel outlier test, the empirical success of which suggests that the failure of existing likelihood-based outlier tests does not necessarily imply "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.13064","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/2010.13064/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":"2010.13064","created_at":"2026-07-05T01:45:51.685293+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.13064v1","created_at":"2026-07-05T01:45:51.685293+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.13064","created_at":"2026-07-05T01:45:51.685293+00:00"},{"alias_kind":"pith_short_12","alias_value":"NJGJI6B3UMRU","created_at":"2026-07-05T01:45:51.685293+00:00"},{"alias_kind":"pith_short_16","alias_value":"NJGJI6B3UMRUZRX4","created_at":"2026-07-05T01:45:51.685293+00:00"},{"alias_kind":"pith_short_8","alias_value":"NJGJI6B3","created_at":"2026-07-05T01:45:51.685293+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.10089","citing_title":"Optimizing Latent Dimension Allocation in Hierarchical VAEs: Balancing Attenuation and Information Retention for OOD Detection","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF","json":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF.json","graph_json":"https://pith.science/api/pith-number/NJGJI6B3UMRUZRX4FUJCFXD6GF/graph.json","events_json":"https://pith.science/api/pith-number/NJGJI6B3UMRUZRX4FUJCFXD6GF/events.json","paper":"https://pith.science/paper/NJGJI6B3"},"agent_actions":{"view_html":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF","download_json":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF.json","view_paper":"https://pith.science/paper/NJGJI6B3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.13064&json=true","fetch_graph":"https://pith.science/api/pith-number/NJGJI6B3UMRUZRX4FUJCFXD6GF/graph.json","fetch_events":"https://pith.science/api/pith-number/NJGJI6B3UMRUZRX4FUJCFXD6GF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF/action/storage_attestation","attest_author":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF/action/author_attestation","sign_citation":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF/action/citation_signature","submit_replication":"https://pith.science/pith/NJGJI6B3UMRUZRX4FUJCFXD6GF/action/replication_record"}},"created_at":"2026-07-05T01:45:51.685293+00:00","updated_at":"2026-07-05T01:45:51.685293+00:00"}