{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:JKFOC7VIL7ILRWTUMKH47TE6HW","short_pith_number":"pith:JKFOC7VI","schema_version":"1.0","canonical_sha256":"4a8ae17ea85fd0b8da74628fcfcc9e3d8900496024b0ca288e991859d46ffa60","source":{"kind":"arxiv","id":"1705.10694","version":3},"attestation_state":"computed","paper":{"title":"Deep Learning is Robust to Massive Label Noise","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.NE"],"primary_cat":"cs.LG","authors_text":"Andreas Veit, David Rolnick, Nir Shavit, Serge Belongie","submitted_at":"2017-05-30T15:10:51Z","abstract_excerpt":"Deep neural networks trained on large supervised datasets have led to impressive results in image classification and other tasks. However, well-annotated datasets can be time-consuming and expensive to collect, lending increased interest to larger but noisy datasets that are more easily obtained. In this paper, we show that deep neural networks are capable of generalizing from training data for which true labels are massively outnumbered by incorrect labels. We demonstrate remarkably high test performance after training on corrupted data from MNIST, CIFAR, and ImageNet. For example, on MNIST w"},"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":"1705.10694","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-30T15:10:51Z","cross_cats_sorted":["cs.AI","cs.CV","cs.NE"],"title_canon_sha256":"609c8b402a9d5eb203ef0b118a372614724925477faac1d74e3a424e2ec5b735","abstract_canon_sha256":"b05e1b8783493bc1a65fa1beaa9f86126aeab92fa382963f96a0b4fcaffe0842"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:22:39.831024Z","signature_b64":"S3dxbj8DJ0TPr0jqnSHAu+rMP3YpnvRygbrLyZ/gA0j6jL1kz+t8sKxoMLQTJWE2D7vhN+wL3em7KdpqJBF5AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a8ae17ea85fd0b8da74628fcfcc9e3d8900496024b0ca288e991859d46ffa60","last_reissued_at":"2026-05-18T00:22:39.830587Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:22:39.830587Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Learning is Robust to Massive Label Noise","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.NE"],"primary_cat":"cs.LG","authors_text":"Andreas Veit, David Rolnick, Nir Shavit, Serge Belongie","submitted_at":"2017-05-30T15:10:51Z","abstract_excerpt":"Deep neural networks trained on large supervised datasets have led to impressive results in image classification and other tasks. However, well-annotated datasets can be time-consuming and expensive to collect, lending increased interest to larger but noisy datasets that are more easily obtained. In this paper, we show that deep neural networks are capable of generalizing from training data for which true labels are massively outnumbered by incorrect labels. We demonstrate remarkably high test performance after training on corrupted data from MNIST, CIFAR, and ImageNet. For example, on MNIST w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.10694","kind":"arxiv","version":3},"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":"1705.10694","created_at":"2026-05-18T00:22:39.830671+00:00"},{"alias_kind":"arxiv_version","alias_value":"1705.10694v3","created_at":"2026-05-18T00:22:39.830671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.10694","created_at":"2026-05-18T00:22:39.830671+00:00"},{"alias_kind":"pith_short_12","alias_value":"JKFOC7VIL7IL","created_at":"2026-05-18T12:31:24.725408+00:00"},{"alias_kind":"pith_short_16","alias_value":"JKFOC7VIL7ILRWTU","created_at":"2026-05-18T12:31:24.725408+00:00"},{"alias_kind":"pith_short_8","alias_value":"JKFOC7VI","created_at":"2026-05-18T12:31:24.725408+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":7,"sample":[{"citing_arxiv_id":"1907.00452","citing_title":"Detecting Spiky Corruption in Markov Decision Processes","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"1907.01742","citing_title":"Supervised Classifiers for Audio Impairments with Noisy Labels","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"1907.11384","citing_title":"Product Image Recognition with Guidance Learning and Noisy Supervision","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2412.00452","citing_title":"Learning Locally, Revising Globally: Global Reviser for Federated Learning with Noisy Labels","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2504.20571","citing_title":"Reinforcement Learning for Reasoning in Large Language Models with One Training Example","ref_index":64,"is_internal_anchor":true},{"citing_arxiv_id":"2603.11749","citing_title":"Truth as a Compression Artifact in Language Model Training","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15032","citing_title":"Multi-Block Attention for Efficient Channel Estimation in IRS-Assisted mmWave MIMO","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2502.02737","citing_title":"SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model","ref_index":217,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21508","citing_title":"BioMiner: A Multi-modal System for Automated Mining of Protein-Ligand Bioactivity Data from Literature","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08051","citing_title":"Inferring Asteroseismic Parameters from Short Observations Using Deep Learning: Application to TESS and K2 Red Giants","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19729","citing_title":"FB-NLL: A Feature-Based Approach to Tackle Noisy Labels in Personalized Federated Learning","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW","json":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW.json","graph_json":"https://pith.science/api/pith-number/JKFOC7VIL7ILRWTUMKH47TE6HW/graph.json","events_json":"https://pith.science/api/pith-number/JKFOC7VIL7ILRWTUMKH47TE6HW/events.json","paper":"https://pith.science/paper/JKFOC7VI"},"agent_actions":{"view_html":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW","download_json":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW.json","view_paper":"https://pith.science/paper/JKFOC7VI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1705.10694&json=true","fetch_graph":"https://pith.science/api/pith-number/JKFOC7VIL7ILRWTUMKH47TE6HW/graph.json","fetch_events":"https://pith.science/api/pith-number/JKFOC7VIL7ILRWTUMKH47TE6HW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW/action/storage_attestation","attest_author":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW/action/author_attestation","sign_citation":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW/action/citation_signature","submit_replication":"https://pith.science/pith/JKFOC7VIL7ILRWTUMKH47TE6HW/action/replication_record"}},"created_at":"2026-05-18T00:22:39.830671+00:00","updated_at":"2026-05-18T00:22:39.830671+00:00"}