{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SFRY5MDRYMPZFSSRZ4YG6L6Q44","short_pith_number":"pith:SFRY5MDR","schema_version":"1.0","canonical_sha256":"91638eb071c31f92ca51cf306f2fd0e71415033719e3c7089a44816817603930","source":{"kind":"arxiv","id":"2211.00453","version":1},"attestation_state":"computed","paper":{"title":"The Perils of Learning From Unlabeled Data: Backdoor Attacks on Semi-supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Amir Houmansadr, Lingjuan Lyu, Virat Shejwalkar","submitted_at":"2022-11-01T13:34:36Z","abstract_excerpt":"Semi-supervised machine learning (SSL) is gaining popularity as it reduces the cost of training ML models. It does so by using very small amounts of (expensive, well-inspected) labeled data and large amounts of (cheap, non-inspected) unlabeled data. SSL has shown comparable or even superior performances compared to conventional fully-supervised ML techniques.\n  In this paper, we show that the key feature of SSL that it can learn from (non-inspected) unlabeled data exposes SSL to strong poisoning attacks. In fact, we argue that, due to its reliance on non-inspected unlabeled data, poisoning is "},"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":"2211.00453","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2022-11-01T13:34:36Z","cross_cats_sorted":[],"title_canon_sha256":"c0c79264ced9c9bf27e766bbbdecafad7e643943945ad9f4c3a076917373d345","abstract_canon_sha256":"4dac27fd8caba19357b7c2330bd6f1f1df6873ab3893b9e84462fe226ccea558"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:22.769603Z","signature_b64":"ESV0RHJ0/6mSsHENpjgIYphfluk8f9fF7RWILo/RFyf8UHor1UCFpL4VN+XzjtPWSUXV38GqwXR7lyCVzu/ZCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91638eb071c31f92ca51cf306f2fd0e71415033719e3c7089a44816817603930","last_reissued_at":"2026-07-05T05:12:22.769132Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:22.769132Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Perils of Learning From Unlabeled Data: Backdoor Attacks on Semi-supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Amir Houmansadr, Lingjuan Lyu, Virat Shejwalkar","submitted_at":"2022-11-01T13:34:36Z","abstract_excerpt":"Semi-supervised machine learning (SSL) is gaining popularity as it reduces the cost of training ML models. It does so by using very small amounts of (expensive, well-inspected) labeled data and large amounts of (cheap, non-inspected) unlabeled data. SSL has shown comparable or even superior performances compared to conventional fully-supervised ML techniques.\n  In this paper, we show that the key feature of SSL that it can learn from (non-inspected) unlabeled data exposes SSL to strong poisoning attacks. In fact, we argue that, due to its reliance on non-inspected unlabeled data, poisoning is "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.00453","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/2211.00453/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":"2211.00453","created_at":"2026-07-05T05:12:22.769186+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.00453v1","created_at":"2026-07-05T05:12:22.769186+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.00453","created_at":"2026-07-05T05:12:22.769186+00:00"},{"alias_kind":"pith_short_12","alias_value":"SFRY5MDRYMPZ","created_at":"2026-07-05T05:12:22.769186+00:00"},{"alias_kind":"pith_short_16","alias_value":"SFRY5MDRYMPZFSSR","created_at":"2026-07-05T05:12:22.769186+00:00"},{"alias_kind":"pith_short_8","alias_value":"SFRY5MDR","created_at":"2026-07-05T05:12:22.769186+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/SFRY5MDRYMPZFSSRZ4YG6L6Q44","json":"https://pith.science/pith/SFRY5MDRYMPZFSSRZ4YG6L6Q44.json","graph_json":"https://pith.science/api/pith-number/SFRY5MDRYMPZFSSRZ4YG6L6Q44/graph.json","events_json":"https://pith.science/api/pith-number/SFRY5MDRYMPZFSSRZ4YG6L6Q44/events.json","paper":"https://pith.science/paper/SFRY5MDR"},"agent_actions":{"view_html":"https://pith.science/pith/SFRY5MDRYMPZFSSRZ4YG6L6Q44","download_json":"https://pith.science/pith/SFRY5MDRYMPZFSSRZ4YG6L6Q44.json","view_paper":"https://pith.science/paper/SFRY5MDR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.00453&json=true","fetch_graph":"https://pith.science/api/pith-number/SFRY5MDRYMPZFSSRZ4YG6L6Q44/graph.json","fetch_events":"https://pith.science/api/pith-number/SFRY5MDRYMPZFSSRZ4YG6L6Q44/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SFRY5MDRYMPZFSSRZ4YG6L6Q44/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SFRY5MDRYMPZFSSRZ4YG6L6Q44/action/storage_attestation","attest_author":"https://pith.science/pith/SFRY5MDRYMPZFSSRZ4YG6L6Q44/action/author_attestation","sign_citation":"https://pith.science/pith/SFRY5MDRYMPZFSSRZ4YG6L6Q44/action/citation_signature","submit_replication":"https://pith.science/pith/SFRY5MDRYMPZFSSRZ4YG6L6Q44/action/replication_record"}},"created_at":"2026-07-05T05:12:22.769186+00:00","updated_at":"2026-07-05T05:12:22.769186+00:00"}