{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:DOAAQPZWFMTJSOVTF6D5LUQEUY","short_pith_number":"pith:DOAAQPZW","schema_version":"1.0","canonical_sha256":"1b80083f362b26993ab32f87d5d204a607933b7f63af2351c1a3b877efc70ae5","source":{"kind":"arxiv","id":"1906.01171","version":2},"attestation_state":"computed","paper":{"title":"Understanding the Limitations of Conditional Generative Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ethan Fetaya, J\\\"orn-Henrik Jacobsen, Richard Zemel, Will Grathwohl","submitted_at":"2019-06-04T02:56:14Z","abstract_excerpt":"Class-conditional generative models hold promise to overcome the shortcomings of their discriminative counterparts. They are a natural choice to solve discriminative tasks in a robust manner as they jointly optimize for predictive performance and accurate modeling of the input distribution. In this work, we investigate robust classification with likelihood-based generative models from a theoretical and practical perspective to investigate if they can deliver on their promises. Our analysis focuses on a spectrum of robustness properties: (1) Detection of worst-case outliers in the form of adver"},"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":"1906.01171","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-04T02:56:14Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"c1d620f172f2eb028c9bcdd0aa99521b05f8bb160261407a7a7de48637dd240b","abstract_canon_sha256":"f19e0ba959a1868aa36e1f7eb3b31cd3aecbc342549ab35c6e7faf63181b1c5a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:40:55.827278Z","signature_b64":"GOJUD5IXU7oKSiVamWnf2JzhkwNHX/xls0woBe33womSJy8i4dakoE/HOOZqkHE64aE5Z6YtKiPFkpRZ5OY9AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b80083f362b26993ab32f87d5d204a607933b7f63af2351c1a3b877efc70ae5","last_reissued_at":"2026-07-05T00:40:55.826800Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:40:55.826800Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding the Limitations of Conditional Generative Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ethan Fetaya, J\\\"orn-Henrik Jacobsen, Richard Zemel, Will Grathwohl","submitted_at":"2019-06-04T02:56:14Z","abstract_excerpt":"Class-conditional generative models hold promise to overcome the shortcomings of their discriminative counterparts. They are a natural choice to solve discriminative tasks in a robust manner as they jointly optimize for predictive performance and accurate modeling of the input distribution. In this work, we investigate robust classification with likelihood-based generative models from a theoretical and practical perspective to investigate if they can deliver on their promises. Our analysis focuses on a spectrum of robustness properties: (1) Detection of worst-case outliers in the form of adver"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.01171","kind":"arxiv","version":2},"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/1906.01171/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":"1906.01171","created_at":"2026-07-05T00:40:55.826859+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.01171v2","created_at":"2026-07-05T00:40:55.826859+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.01171","created_at":"2026-07-05T00:40:55.826859+00:00"},{"alias_kind":"pith_short_12","alias_value":"DOAAQPZWFMTJ","created_at":"2026-07-05T00:40:55.826859+00:00"},{"alias_kind":"pith_short_16","alias_value":"DOAAQPZWFMTJSOVT","created_at":"2026-07-05T00:40:55.826859+00:00"},{"alias_kind":"pith_short_8","alias_value":"DOAAQPZW","created_at":"2026-07-05T00:40:55.826859+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.20498","citing_title":"ControlTac: Force- and Position-Controlled Tactile Data Augmentation with a Single Reference Image","ref_index":61,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY","json":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY.json","graph_json":"https://pith.science/api/pith-number/DOAAQPZWFMTJSOVTF6D5LUQEUY/graph.json","events_json":"https://pith.science/api/pith-number/DOAAQPZWFMTJSOVTF6D5LUQEUY/events.json","paper":"https://pith.science/paper/DOAAQPZW"},"agent_actions":{"view_html":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY","download_json":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY.json","view_paper":"https://pith.science/paper/DOAAQPZW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.01171&json=true","fetch_graph":"https://pith.science/api/pith-number/DOAAQPZWFMTJSOVTF6D5LUQEUY/graph.json","fetch_events":"https://pith.science/api/pith-number/DOAAQPZWFMTJSOVTF6D5LUQEUY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY/action/storage_attestation","attest_author":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY/action/author_attestation","sign_citation":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY/action/citation_signature","submit_replication":"https://pith.science/pith/DOAAQPZWFMTJSOVTF6D5LUQEUY/action/replication_record"}},"created_at":"2026-07-05T00:40:55.826859+00:00","updated_at":"2026-07-05T00:40:55.826859+00:00"}