{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:26O3HXHHSLMSKAR3YK2AFZ5K5T","short_pith_number":"pith:26O3HXHH","schema_version":"1.0","canonical_sha256":"d79db3dce792d925023bc2b402e7aaecc158dba5076919c0c9dd8bd91347a0a5","source":{"kind":"arxiv","id":"2104.09563","version":1},"attestation_state":"computed","paper":{"title":"A Framework using Contrastive Learning for Classification with Noisy Labels","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Madalina Ciortan, Romain Dupuis, Thomas Peel","submitted_at":"2021-04-19T18:51:22Z","abstract_excerpt":"We propose a framework using contrastive learning as a pre-training task to perform image classification in the presence of noisy labels. Recent strategies such as pseudo-labeling, sample selection with Gaussian Mixture models, weighted supervised contrastive learning have been combined into a fine-tuning phase following the pre-training. This paper provides an extensive empirical study showing that a preliminary contrastive learning step brings a significant gain in performance when using different loss functions: non-robust, robust, and early-learning regularized. Our experiments performed o"},"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":"2104.09563","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-04-19T18:51:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"976457084ac3f9d8bb84a6728d0c8f144e801ea880ad9259431541b415173717","abstract_canon_sha256":"733e9f1036fbe47869dafcd54e78b0757f038feac7d2c1b41d2c75bcc5bc75c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:33:34.874943Z","signature_b64":"6YiBCk6NlnSkJ/4cwckztQFQtY/SVyZlBq0DtfuSvlnXH8+z31cLqjoSTFmTn2J7ViP5Gx2CZ7hmdt1k5y6sCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d79db3dce792d925023bc2b402e7aaecc158dba5076919c0c9dd8bd91347a0a5","last_reissued_at":"2026-07-05T02:33:34.874354Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:33:34.874354Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Framework using Contrastive Learning for Classification with Noisy Labels","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Madalina Ciortan, Romain Dupuis, Thomas Peel","submitted_at":"2021-04-19T18:51:22Z","abstract_excerpt":"We propose a framework using contrastive learning as a pre-training task to perform image classification in the presence of noisy labels. Recent strategies such as pseudo-labeling, sample selection with Gaussian Mixture models, weighted supervised contrastive learning have been combined into a fine-tuning phase following the pre-training. This paper provides an extensive empirical study showing that a preliminary contrastive learning step brings a significant gain in performance when using different loss functions: non-robust, robust, and early-learning regularized. Our experiments performed o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.09563","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/2104.09563/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":"2104.09563","created_at":"2026-07-05T02:33:34.874430+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.09563v1","created_at":"2026-07-05T02:33:34.874430+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.09563","created_at":"2026-07-05T02:33:34.874430+00:00"},{"alias_kind":"pith_short_12","alias_value":"26O3HXHHSLMS","created_at":"2026-07-05T02:33:34.874430+00:00"},{"alias_kind":"pith_short_16","alias_value":"26O3HXHHSLMSKAR3","created_at":"2026-07-05T02:33:34.874430+00:00"},{"alias_kind":"pith_short_8","alias_value":"26O3HXHH","created_at":"2026-07-05T02:33:34.874430+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T","json":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T.json","graph_json":"https://pith.science/api/pith-number/26O3HXHHSLMSKAR3YK2AFZ5K5T/graph.json","events_json":"https://pith.science/api/pith-number/26O3HXHHSLMSKAR3YK2AFZ5K5T/events.json","paper":"https://pith.science/paper/26O3HXHH"},"agent_actions":{"view_html":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T","download_json":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T.json","view_paper":"https://pith.science/paper/26O3HXHH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.09563&json=true","fetch_graph":"https://pith.science/api/pith-number/26O3HXHHSLMSKAR3YK2AFZ5K5T/graph.json","fetch_events":"https://pith.science/api/pith-number/26O3HXHHSLMSKAR3YK2AFZ5K5T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T/action/storage_attestation","attest_author":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T/action/author_attestation","sign_citation":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T/action/citation_signature","submit_replication":"https://pith.science/pith/26O3HXHHSLMSKAR3YK2AFZ5K5T/action/replication_record"}},"created_at":"2026-07-05T02:33:34.874430+00:00","updated_at":"2026-07-05T02:33:34.874430+00:00"}