{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:P6PBE3MV34BCO3GPHJ22SPDO3F","short_pith_number":"pith:P6PBE3MV","schema_version":"1.0","canonical_sha256":"7f9e126d95df02276ccf3a75a93c6ed959a70a27c18c8cf4a031bb0594eaa32e","source":{"kind":"arxiv","id":"1901.09960","version":5},"attestation_state":"computed","paper":{"title":"Using Pre-Training Can Improve Model Robustness and Uncertainty","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dan Hendrycks, Kimin Lee, Mantas Mazeika","submitted_at":"2019-01-28T19:37:07Z","abstract_excerpt":"He et al. (2018) have called into question the utility of pre-training by showing that training from scratch can often yield similar performance to pre-training. We show that although pre-training may not improve performance on traditional classification metrics, it improves model robustness and uncertainty estimates. Through extensive experiments on adversarial examples, label corruption, class imbalance, out-of-distribution detection, and confidence calibration, we demonstrate large gains from pre-training and complementary effects with task-specific methods. We introduce adversarial pre-tra"},"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":"1901.09960","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-28T19:37:07Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"d7f0a3eccad003a10ecca88841afc36dbb779efffbfe2a178e98fbfe9ea16dd2","abstract_canon_sha256":"a263532117c50906498ea4d9abb32d6e002b082c462823f41c37703a2be29c03"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:13:13.361392Z","signature_b64":"jAW0IWv87UsPxKPxJmdUKqy/LYikBkwkcN6LB9MeBhCapFfRgOJ83R2zgG7aBYrC6NGMXWo/bvLLMfp8gmqsAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f9e126d95df02276ccf3a75a93c6ed959a70a27c18c8cf4a031bb0594eaa32e","last_reissued_at":"2026-07-05T00:13:13.361000Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:13:13.361000Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Using Pre-Training Can Improve Model Robustness and Uncertainty","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dan Hendrycks, Kimin Lee, Mantas Mazeika","submitted_at":"2019-01-28T19:37:07Z","abstract_excerpt":"He et al. (2018) have called into question the utility of pre-training by showing that training from scratch can often yield similar performance to pre-training. We show that although pre-training may not improve performance on traditional classification metrics, it improves model robustness and uncertainty estimates. Through extensive experiments on adversarial examples, label corruption, class imbalance, out-of-distribution detection, and confidence calibration, we demonstrate large gains from pre-training and complementary effects with task-specific methods. We introduce adversarial pre-tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.09960","kind":"arxiv","version":5},"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/1901.09960/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":"1901.09960","created_at":"2026-07-05T00:13:13.361055+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.09960v5","created_at":"2026-07-05T00:13:13.361055+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.09960","created_at":"2026-07-05T00:13:13.361055+00:00"},{"alias_kind":"pith_short_12","alias_value":"P6PBE3MV34BC","created_at":"2026-07-05T00:13:13.361055+00:00"},{"alias_kind":"pith_short_16","alias_value":"P6PBE3MV34BCO3GP","created_at":"2026-07-05T00:13:13.361055+00:00"},{"alias_kind":"pith_short_8","alias_value":"P6PBE3MV","created_at":"2026-07-05T00:13:13.361055+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.11400","citing_title":"The model is the message: Lightweight convolutional autoencoders applied to noisy imaging data for planetary science and astrobiology","ref_index":26,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F","json":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F.json","graph_json":"https://pith.science/api/pith-number/P6PBE3MV34BCO3GPHJ22SPDO3F/graph.json","events_json":"https://pith.science/api/pith-number/P6PBE3MV34BCO3GPHJ22SPDO3F/events.json","paper":"https://pith.science/paper/P6PBE3MV"},"agent_actions":{"view_html":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F","download_json":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F.json","view_paper":"https://pith.science/paper/P6PBE3MV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.09960&json=true","fetch_graph":"https://pith.science/api/pith-number/P6PBE3MV34BCO3GPHJ22SPDO3F/graph.json","fetch_events":"https://pith.science/api/pith-number/P6PBE3MV34BCO3GPHJ22SPDO3F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F/action/storage_attestation","attest_author":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F/action/author_attestation","sign_citation":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F/action/citation_signature","submit_replication":"https://pith.science/pith/P6PBE3MV34BCO3GPHJ22SPDO3F/action/replication_record"}},"created_at":"2026-07-05T00:13:13.361055+00:00","updated_at":"2026-07-05T00:13:13.361055+00:00"}