{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:YWBUC4RDDAMPHO6Z2G3JESM5H2","short_pith_number":"pith:YWBUC4RD","schema_version":"1.0","canonical_sha256":"c5834172231818f3bbd9d1b692499d3ea8c30450afc62f5e51b973daaa023e5b","source":{"kind":"arxiv","id":"2206.05259","version":1},"attestation_state":"computed","paper":{"title":"Is Self-Supervised Learning More Robust Than Supervised Learning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Haoran Tang, Jian Peng, Junkun Chen, Yuanyi Zhong, Yu-Xiong Wang","submitted_at":"2022-06-10T17:58:00Z","abstract_excerpt":"Self-supervised contrastive learning is a powerful tool to learn visual representation without labels. Prior work has primarily focused on evaluating the recognition accuracy of various pre-training algorithms, but has overlooked other behavioral aspects. In addition to accuracy, distributional robustness plays a critical role in the reliability of machine learning models. We design and conduct a series of robustness tests to quantify the behavioral differences between contrastive learning and supervised learning to downstream or pre-training data distribution changes. These tests leverage dat"},"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":"2206.05259","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-10T17:58:00Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d8fc891b89d970291bdd11bdf610500124a57668a03b10d276e3658ade1e078d","abstract_canon_sha256":"08fdd89fed4c2a419ed6d9aa550978c9bfca728bdd1f6ae487adb92fe21656e0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:30:48.906456Z","signature_b64":"OvyE9Z3YD8OnwaK3MY/vBvgCsS/UzjF9DMAsPc13gdgm+ub155UICmMQo5Hps4nZ5hEvCTJS3YSTmfq2Wq1oBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c5834172231818f3bbd9d1b692499d3ea8c30450afc62f5e51b973daaa023e5b","last_reissued_at":"2026-07-05T04:30:48.906072Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:30:48.906072Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is Self-Supervised Learning More Robust Than Supervised Learning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Haoran Tang, Jian Peng, Junkun Chen, Yuanyi Zhong, Yu-Xiong Wang","submitted_at":"2022-06-10T17:58:00Z","abstract_excerpt":"Self-supervised contrastive learning is a powerful tool to learn visual representation without labels. Prior work has primarily focused on evaluating the recognition accuracy of various pre-training algorithms, but has overlooked other behavioral aspects. In addition to accuracy, distributional robustness plays a critical role in the reliability of machine learning models. We design and conduct a series of robustness tests to quantify the behavioral differences between contrastive learning and supervised learning to downstream or pre-training data distribution changes. These tests leverage dat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.05259","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/2206.05259/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":"2206.05259","created_at":"2026-07-05T04:30:48.906123+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.05259v1","created_at":"2026-07-05T04:30:48.906123+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.05259","created_at":"2026-07-05T04:30:48.906123+00:00"},{"alias_kind":"pith_short_12","alias_value":"YWBUC4RDDAMP","created_at":"2026-07-05T04:30:48.906123+00:00"},{"alias_kind":"pith_short_16","alias_value":"YWBUC4RDDAMPHO6Z","created_at":"2026-07-05T04:30:48.906123+00:00"},{"alias_kind":"pith_short_8","alias_value":"YWBUC4RD","created_at":"2026-07-05T04:30:48.906123+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.10351","citing_title":"A Unified and Scalable Membership Inference Method for Visual Self-supervised Encoder via Part-aware Capability","ref_index":83,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2","json":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2.json","graph_json":"https://pith.science/api/pith-number/YWBUC4RDDAMPHO6Z2G3JESM5H2/graph.json","events_json":"https://pith.science/api/pith-number/YWBUC4RDDAMPHO6Z2G3JESM5H2/events.json","paper":"https://pith.science/paper/YWBUC4RD"},"agent_actions":{"view_html":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2","download_json":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2.json","view_paper":"https://pith.science/paper/YWBUC4RD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.05259&json=true","fetch_graph":"https://pith.science/api/pith-number/YWBUC4RDDAMPHO6Z2G3JESM5H2/graph.json","fetch_events":"https://pith.science/api/pith-number/YWBUC4RDDAMPHO6Z2G3JESM5H2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2/action/storage_attestation","attest_author":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2/action/author_attestation","sign_citation":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2/action/citation_signature","submit_replication":"https://pith.science/pith/YWBUC4RDDAMPHO6Z2G3JESM5H2/action/replication_record"}},"created_at":"2026-07-05T04:30:48.906123+00:00","updated_at":"2026-07-05T04:30:48.906123+00:00"}