{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:G3SCX23D7BZJUP7MZPR3ENTDF7","short_pith_number":"pith:G3SCX23D","schema_version":"1.0","canonical_sha256":"36e42beb63f8729a3feccbe3b236632fd0932463ee8f99b3bf919cfea99dab67","source":{"kind":"arxiv","id":"2406.11544","version":4},"attestation_state":"computed","paper":{"title":"Do Parameters Reveal More than Loss for Membership Inference?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Anshuman Suri, David Evans, Xiao Zhang","submitted_at":"2024-06-17T13:42:28Z","abstract_excerpt":"Membership inference attacks are used as a key tool for disclosure auditing. They aim to infer whether an individual record was used to train a model. While such evaluations are useful to demonstrate risk, they are computationally expensive and often make strong assumptions about potential adversaries' access to models and training environments, and thus do not provide tight bounds on leakage from potential attacks. We show how prior claims around black-box access being sufficient for optimal membership inference do not hold for stochastic gradient descent, and that optimal membership inferenc"},"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":"2406.11544","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-06-17T13:42:28Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"a1c282a861ba86d81e174e5d8423419932bdbe39ad293e7977baa7ab4bea389c","abstract_canon_sha256":"15265ad3d34109fa5b34c3028e0f8a3758e7e5e808062bea2c315e16ab4cc6e2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:51:33.625182Z","signature_b64":"8yBEmYPzZH9jBzYM5zUzo7hxImm4rS2jSe81MlFJobaIKRIFOGLhA1WZTOehrrYyvt7AeZAXcvpPeQGdf4iFBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"36e42beb63f8729a3feccbe3b236632fd0932463ee8f99b3bf919cfea99dab67","last_reissued_at":"2026-07-05T09:51:33.624656Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:51:33.624656Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do Parameters Reveal More than Loss for Membership Inference?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Anshuman Suri, David Evans, Xiao Zhang","submitted_at":"2024-06-17T13:42:28Z","abstract_excerpt":"Membership inference attacks are used as a key tool for disclosure auditing. They aim to infer whether an individual record was used to train a model. While such evaluations are useful to demonstrate risk, they are computationally expensive and often make strong assumptions about potential adversaries' access to models and training environments, and thus do not provide tight bounds on leakage from potential attacks. We show how prior claims around black-box access being sufficient for optimal membership inference do not hold for stochastic gradient descent, and that optimal membership inferenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.11544","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/2406.11544/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":"2406.11544","created_at":"2026-07-05T09:51:33.624715+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.11544v4","created_at":"2026-07-05T09:51:33.624715+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.11544","created_at":"2026-07-05T09:51:33.624715+00:00"},{"alias_kind":"pith_short_12","alias_value":"G3SCX23D7BZJ","created_at":"2026-07-05T09:51:33.624715+00:00"},{"alias_kind":"pith_short_16","alias_value":"G3SCX23D7BZJUP7M","created_at":"2026-07-05T09:51:33.624715+00:00"},{"alias_kind":"pith_short_8","alias_value":"G3SCX23D","created_at":"2026-07-05T09:51:33.624715+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.00306","citing_title":"Riddle Me This! Stealthy Membership Inference for Retrieval-Augmented Generation","ref_index":52,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7","json":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7.json","graph_json":"https://pith.science/api/pith-number/G3SCX23D7BZJUP7MZPR3ENTDF7/graph.json","events_json":"https://pith.science/api/pith-number/G3SCX23D7BZJUP7MZPR3ENTDF7/events.json","paper":"https://pith.science/paper/G3SCX23D"},"agent_actions":{"view_html":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7","download_json":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7.json","view_paper":"https://pith.science/paper/G3SCX23D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.11544&json=true","fetch_graph":"https://pith.science/api/pith-number/G3SCX23D7BZJUP7MZPR3ENTDF7/graph.json","fetch_events":"https://pith.science/api/pith-number/G3SCX23D7BZJUP7MZPR3ENTDF7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7/action/storage_attestation","attest_author":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7/action/author_attestation","sign_citation":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7/action/citation_signature","submit_replication":"https://pith.science/pith/G3SCX23D7BZJUP7MZPR3ENTDF7/action/replication_record"}},"created_at":"2026-07-05T09:51:33.624715+00:00","updated_at":"2026-07-05T09:51:33.624715+00:00"}