{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:LUPKECKA4WVHYGXJAYKHWVYY3Q","short_pith_number":"pith:LUPKECKA","schema_version":"1.0","canonical_sha256":"5d1ea20940e5aa7c1ae906147b5718dc1e91c44b9c2cceee7a87787692ae76c1","source":{"kind":"arxiv","id":"1905.04899","version":2},"attestation_state":"computed","paper":{"title":"CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Dongyoon Han, Junsuk Choe, Sangdoo Yun, Sanghyuk Chun, Seong Joon Oh, Youngjoon Yoo","submitted_at":"2019-05-13T08:10:22Z","abstract_excerpt":"Regional dropout strategies have been proposed to enhance the performance of convolutional neural network classifiers. They have proved to be effective for guiding the model to attend on less discriminative parts of objects (e.g. leg as opposed to head of a person), thereby letting the network generalize better and have better object localization capabilities. On the other hand, current methods for regional dropout remove informative pixels on training images by overlaying a patch of either black pixels or random noise. Such removal is not desirable because it leads to information loss and ine"},"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":"1905.04899","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-05-13T08:10:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"05d70c0665151b172552a19bf9126028b29c5ab6fe38441854b3b1aeef2b95eb","abstract_canon_sha256":"423dcbe8a01a326ebdbc75be3aed0716fad6fccfba15c8cf79a90979cb0266a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:52:29.269016Z","signature_b64":"Q8YAE7yZ+HAdQPXW7Y/4UYniMxmVTjLZtArZJhUn4/f0xFSOBflldXB3TpAXz49GrJEm+ixQ5xx24JoIMuyHAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5d1ea20940e5aa7c1ae906147b5718dc1e91c44b9c2cceee7a87787692ae76c1","last_reissued_at":"2026-07-04T23:52:29.268532Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:52:29.268532Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Dongyoon Han, Junsuk Choe, Sangdoo Yun, Sanghyuk Chun, Seong Joon Oh, Youngjoon Yoo","submitted_at":"2019-05-13T08:10:22Z","abstract_excerpt":"Regional dropout strategies have been proposed to enhance the performance of convolutional neural network classifiers. They have proved to be effective for guiding the model to attend on less discriminative parts of objects (e.g. leg as opposed to head of a person), thereby letting the network generalize better and have better object localization capabilities. On the other hand, current methods for regional dropout remove informative pixels on training images by overlaying a patch of either black pixels or random noise. Such removal is not desirable because it leads to information loss and ine"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.04899","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/1905.04899/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":"1905.04899","created_at":"2026-07-04T23:52:29.268591+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.04899v2","created_at":"2026-07-04T23:52:29.268591+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.04899","created_at":"2026-07-04T23:52:29.268591+00:00"},{"alias_kind":"pith_short_12","alias_value":"LUPKECKA4WVH","created_at":"2026-07-04T23:52:29.268591+00:00"},{"alias_kind":"pith_short_16","alias_value":"LUPKECKA4WVHYGXJ","created_at":"2026-07-04T23:52:29.268591+00:00"},{"alias_kind":"pith_short_8","alias_value":"LUPKECKA","created_at":"2026-07-04T23:52:29.268591+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.03803","citing_title":"R\\'enyi Attention Entropy for Patch Pruning","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09716","citing_title":"Medical Model Synthesis Architectures: A Case Study","ref_index":176,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00652","citing_title":"D-SHIFT: Transferring High Spatial Information from GRACE Monthly TWSA Mascon to Daily Products Using Generative Adversarial Networks","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06678","citing_title":"A Wasserstein GAN-based climate scenario generator for risk management and insurance: the case of soil subsidence","ref_index":66,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q","json":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q.json","graph_json":"https://pith.science/api/pith-number/LUPKECKA4WVHYGXJAYKHWVYY3Q/graph.json","events_json":"https://pith.science/api/pith-number/LUPKECKA4WVHYGXJAYKHWVYY3Q/events.json","paper":"https://pith.science/paper/LUPKECKA"},"agent_actions":{"view_html":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q","download_json":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q.json","view_paper":"https://pith.science/paper/LUPKECKA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.04899&json=true","fetch_graph":"https://pith.science/api/pith-number/LUPKECKA4WVHYGXJAYKHWVYY3Q/graph.json","fetch_events":"https://pith.science/api/pith-number/LUPKECKA4WVHYGXJAYKHWVYY3Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q/action/storage_attestation","attest_author":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q/action/author_attestation","sign_citation":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q/action/citation_signature","submit_replication":"https://pith.science/pith/LUPKECKA4WVHYGXJAYKHWVYY3Q/action/replication_record"}},"created_at":"2026-07-04T23:52:29.268591+00:00","updated_at":"2026-07-04T23:52:29.268591+00:00"}