{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:A2LWOUA3HTLOXPVWR6AYHU5CE4","short_pith_number":"pith:A2LWOUA3","schema_version":"1.0","canonical_sha256":"069767501b3cd6ebbeb68f8183d3a22739a72b590dcd461080766938746550ef","source":{"kind":"arxiv","id":"1906.02611","version":1},"attestation_state":"computed","paper":{"title":"Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ben Poole, Dong Yin, Ekin D. Cubuk, Justin Gilmer, Raphael Gontijo Lopes","submitted_at":"2019-06-06T17:54:24Z","abstract_excerpt":"Deploying machine learning systems in the real world requires both high accuracy on clean data and robustness to naturally occurring corruptions. While architectural advances have led to improved accuracy, building robust models remains challenging. Prior work has argued that there is an inherent trade-off between robustness and accuracy, which is exemplified by standard data augment techniques such as Cutout, which improves clean accuracy but not robustness, and additive Gaussian noise, which improves robustness but hurts accuracy. To overcome this trade-off, we introduce Patch Gaussian, a si"},"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":"1906.02611","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-06T17:54:24Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"4e101fe14403010e566071c1a0993c08df3b8e29c00cc028303d7f7c30b74afd","abstract_canon_sha256":"7dc1d3b900ebe3ebb6da545fd64c03a6320370cc45eb408cf500dcbe1dd1d800"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:44:00.604358Z","signature_b64":"RbtbC55Cr8IObGcwmcVNy9qb8V9ZVvo/cE7FoBAq8RMvm1yIKBxZmxmZagItuuWz28wOkYgtotgLifYuoxZzBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"069767501b3cd6ebbeb68f8183d3a22739a72b590dcd461080766938746550ef","last_reissued_at":"2026-05-17T23:44:00.603647Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:44:00.603647Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ben Poole, Dong Yin, Ekin D. Cubuk, Justin Gilmer, Raphael Gontijo Lopes","submitted_at":"2019-06-06T17:54:24Z","abstract_excerpt":"Deploying machine learning systems in the real world requires both high accuracy on clean data and robustness to naturally occurring corruptions. While architectural advances have led to improved accuracy, building robust models remains challenging. Prior work has argued that there is an inherent trade-off between robustness and accuracy, which is exemplified by standard data augment techniques such as Cutout, which improves clean accuracy but not robustness, and additive Gaussian noise, which improves robustness but hurts accuracy. To overcome this trade-off, we introduce Patch Gaussian, a si"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.02611","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":""},"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":"1906.02611","created_at":"2026-05-17T23:44:00.603760+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.02611v1","created_at":"2026-05-17T23:44:00.603760+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.02611","created_at":"2026-05-17T23:44:00.603760+00:00"},{"alias_kind":"pith_short_12","alias_value":"A2LWOUA3HTLO","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_16","alias_value":"A2LWOUA3HTLOXPVW","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_8","alias_value":"A2LWOUA3","created_at":"2026-05-18T12:33:12.712433+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.15306","citing_title":"How Data Augmentation Shapes Neural Representations","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17451","citing_title":"SegTTA: Training-Free Test-Time Augmentation for Zero-Shot Medical Imaging Segmentation","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4","json":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4.json","graph_json":"https://pith.science/api/pith-number/A2LWOUA3HTLOXPVWR6AYHU5CE4/graph.json","events_json":"https://pith.science/api/pith-number/A2LWOUA3HTLOXPVWR6AYHU5CE4/events.json","paper":"https://pith.science/paper/A2LWOUA3"},"agent_actions":{"view_html":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4","download_json":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4.json","view_paper":"https://pith.science/paper/A2LWOUA3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.02611&json=true","fetch_graph":"https://pith.science/api/pith-number/A2LWOUA3HTLOXPVWR6AYHU5CE4/graph.json","fetch_events":"https://pith.science/api/pith-number/A2LWOUA3HTLOXPVWR6AYHU5CE4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4/action/storage_attestation","attest_author":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4/action/author_attestation","sign_citation":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4/action/citation_signature","submit_replication":"https://pith.science/pith/A2LWOUA3HTLOXPVWR6AYHU5CE4/action/replication_record"}},"created_at":"2026-05-17T23:44:00.603760+00:00","updated_at":"2026-05-17T23:44:00.603760+00:00"}