{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:IWTXWHPRUD4WW4M4BBKKHLM4JS","short_pith_number":"pith:IWTXWHPR","schema_version":"1.0","canonical_sha256":"45a77b1df1a0f96b719c0854a3ad9c4c9efc679cc6add7028adeb353aaf1a8a6","source":{"kind":"arxiv","id":"2007.14321","version":3},"attestation_state":"computed","paper":{"title":"Label-Only Membership Inference Attacks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CR","authors_text":"Christopher A. Choquette-Choo, Florian Tramer, Nicholas Carlini, Nicolas Papernot","submitted_at":"2020-07-28T15:44:31Z","abstract_excerpt":"Membership inference attacks are one of the simplest forms of privacy leakage for machine learning models: given a data point and model, determine whether the point was used to train the model. Existing membership inference attacks exploit models' abnormal confidence when queried on their training data. These attacks do not apply if the adversary only gets access to models' predicted labels, without a confidence measure. In this paper, we introduce label-only membership inference attacks. Instead of relying on confidence scores, our attacks evaluate the robustness of a model's predicted labels"},"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":"2007.14321","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2020-07-28T15:44:31Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"a699a3fcd73cc3ccbbe0a33e56eb9bdd5105cb6ab6f35990e714a00125f0ca66","abstract_canon_sha256":"b62b3dca880f7b8706dab36cca8de75b0f096c34a3d2474242e67f5edf8f0eaf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:37:38.318740Z","signature_b64":"5UkVEWMr9iKOP4p1MpuykKsvGbRXaR2eLMPbUqi4o49pooJjdrsi5lt+lOohzH+yKqOP9/M06IQlmdHZBmOvBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"45a77b1df1a0f96b719c0854a3ad9c4c9efc679cc6add7028adeb353aaf1a8a6","last_reissued_at":"2026-07-05T03:37:38.318198Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:37:38.318198Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Label-Only Membership Inference Attacks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CR","authors_text":"Christopher A. Choquette-Choo, Florian Tramer, Nicholas Carlini, Nicolas Papernot","submitted_at":"2020-07-28T15:44:31Z","abstract_excerpt":"Membership inference attacks are one of the simplest forms of privacy leakage for machine learning models: given a data point and model, determine whether the point was used to train the model. Existing membership inference attacks exploit models' abnormal confidence when queried on their training data. These attacks do not apply if the adversary only gets access to models' predicted labels, without a confidence measure. In this paper, we introduce label-only membership inference attacks. Instead of relying on confidence scores, our attacks evaluate the robustness of a model's predicted labels"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.14321","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2007.14321/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":"2007.14321","created_at":"2026-07-05T03:37:38.318259+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.14321v3","created_at":"2026-07-05T03:37:38.318259+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.14321","created_at":"2026-07-05T03:37:38.318259+00:00"},{"alias_kind":"pith_short_12","alias_value":"IWTXWHPRUD4W","created_at":"2026-07-05T03:37:38.318259+00:00"},{"alias_kind":"pith_short_16","alias_value":"IWTXWHPRUD4WW4M4","created_at":"2026-07-05T03:37:38.318259+00:00"},{"alias_kind":"pith_short_8","alias_value":"IWTXWHPR","created_at":"2026-07-05T03:37:38.318259+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30462","citing_title":"idSCD: Identifying Training Datasets through Semantic Correlation Descriptors","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS","json":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS.json","graph_json":"https://pith.science/api/pith-number/IWTXWHPRUD4WW4M4BBKKHLM4JS/graph.json","events_json":"https://pith.science/api/pith-number/IWTXWHPRUD4WW4M4BBKKHLM4JS/events.json","paper":"https://pith.science/paper/IWTXWHPR"},"agent_actions":{"view_html":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS","download_json":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS.json","view_paper":"https://pith.science/paper/IWTXWHPR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.14321&json=true","fetch_graph":"https://pith.science/api/pith-number/IWTXWHPRUD4WW4M4BBKKHLM4JS/graph.json","fetch_events":"https://pith.science/api/pith-number/IWTXWHPRUD4WW4M4BBKKHLM4JS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS/action/storage_attestation","attest_author":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS/action/author_attestation","sign_citation":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS/action/citation_signature","submit_replication":"https://pith.science/pith/IWTXWHPRUD4WW4M4BBKKHLM4JS/action/replication_record"}},"created_at":"2026-07-05T03:37:38.318259+00:00","updated_at":"2026-07-05T03:37:38.318259+00:00"}