{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:OOMMZUHTUETXETTHM4MDKQAYN6","short_pith_number":"pith:OOMMZUHT","schema_version":"1.0","canonical_sha256":"7398ccd0f3a127724e6767183540186f918d042508bea64fe4900721a2b8cf97","source":{"kind":"arxiv","id":"2003.10580","version":4},"attestation_state":"computed","paper":{"title":"Meta Pseudo Labels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hieu Pham, Minh-Thang Luong, Qizhe Xie, Quoc V. Le, Zihang Dai","submitted_at":"2020-03-23T23:41:57Z","abstract_excerpt":"We present Meta Pseudo Labels, a semi-supervised learning method that achieves a new state-of-the-art top-1 accuracy of 90.2% on ImageNet, which is 1.6% better than the existing state-of-the-art. Like Pseudo Labels, Meta Pseudo Labels has a teacher network to generate pseudo labels on unlabeled data to teach a student network. However, unlike Pseudo Labels where the teacher is fixed, the teacher in Meta Pseudo Labels is constantly adapted by the feedback of the student's performance on the labeled dataset. As a result, the teacher generates better pseudo labels to teach the student. Our code 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":"2003.10580","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-03-23T23:41:57Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"3bb26f13d45835c46d2dbdda56d6b1611bf2a2edb473dfdaeb361009d2786fd2","abstract_canon_sha256":"0f781d318cc508259c37242715230fc57f738688faf6b02226d6c87b235db01a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:19:14.753983Z","signature_b64":"NVDmequNi2ZO8mrdjx7DyRezsHg3LCO2ql69pbUNXB9EDoyoG4UZo4OxpfHfm5w/YfGok+Rm0zmg2MEbG8vnCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7398ccd0f3a127724e6767183540186f918d042508bea64fe4900721a2b8cf97","last_reissued_at":"2026-07-05T02:19:14.753503Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:19:14.753503Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Meta Pseudo Labels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hieu Pham, Minh-Thang Luong, Qizhe Xie, Quoc V. Le, Zihang Dai","submitted_at":"2020-03-23T23:41:57Z","abstract_excerpt":"We present Meta Pseudo Labels, a semi-supervised learning method that achieves a new state-of-the-art top-1 accuracy of 90.2% on ImageNet, which is 1.6% better than the existing state-of-the-art. Like Pseudo Labels, Meta Pseudo Labels has a teacher network to generate pseudo labels on unlabeled data to teach a student network. However, unlike Pseudo Labels where the teacher is fixed, the teacher in Meta Pseudo Labels is constantly adapted by the feedback of the student's performance on the labeled dataset. As a result, the teacher generates better pseudo labels to teach the student. Our code w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.10580","kind":"arxiv","version":4},"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/2003.10580/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":"2003.10580","created_at":"2026-07-05T02:19:14.753573+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.10580v4","created_at":"2026-07-05T02:19:14.753573+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.10580","created_at":"2026-07-05T02:19:14.753573+00:00"},{"alias_kind":"pith_short_12","alias_value":"OOMMZUHTUETX","created_at":"2026-07-05T02:19:14.753573+00:00"},{"alias_kind":"pith_short_16","alias_value":"OOMMZUHTUETXETTH","created_at":"2026-07-05T02:19:14.753573+00:00"},{"alias_kind":"pith_short_8","alias_value":"OOMMZUHT","created_at":"2026-07-05T02:19:14.753573+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2104.14294","citing_title":"Emerging Properties in Self-Supervised Vision Transformers","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16588","citing_title":"Vision Transformers Need Registers","ref_index":179,"is_internal_anchor":false},{"citing_arxiv_id":"2404.08471","citing_title":"Revisiting Feature Prediction for Learning Visual Representations from Video","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6","json":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6.json","graph_json":"https://pith.science/api/pith-number/OOMMZUHTUETXETTHM4MDKQAYN6/graph.json","events_json":"https://pith.science/api/pith-number/OOMMZUHTUETXETTHM4MDKQAYN6/events.json","paper":"https://pith.science/paper/OOMMZUHT"},"agent_actions":{"view_html":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6","download_json":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6.json","view_paper":"https://pith.science/paper/OOMMZUHT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.10580&json=true","fetch_graph":"https://pith.science/api/pith-number/OOMMZUHTUETXETTHM4MDKQAYN6/graph.json","fetch_events":"https://pith.science/api/pith-number/OOMMZUHTUETXETTHM4MDKQAYN6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6/action/storage_attestation","attest_author":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6/action/author_attestation","sign_citation":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6/action/citation_signature","submit_replication":"https://pith.science/pith/OOMMZUHTUETXETTHM4MDKQAYN6/action/replication_record"}},"created_at":"2026-07-05T02:19:14.753573+00:00","updated_at":"2026-07-05T02:19:14.753573+00:00"}