{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:5UXZVVEIIZMPOJCJYFUALO6ZS7","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d72cc5b3fece72928462c260e5d5284ab71ce0d2017b0b84f49c7118a17a6318","cross_cats_sorted":["cs.CL","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-31T15:44:43Z","title_canon_sha256":"cf9def1fb7be33597019208bbe14514be554560c1ed944f450deef9c5a77ec71"},"schema_version":"1.0","source":{"id":"2608.00144","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.00144","created_at":"2026-08-04T00:33:03Z"},{"alias_kind":"arxiv_version","alias_value":"2608.00144v1","created_at":"2026-08-04T00:33:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00144","created_at":"2026-08-04T00:33:03Z"},{"alias_kind":"pith_short_12","alias_value":"5UXZVVEIIZMP","created_at":"2026-08-04T00:33:03Z"},{"alias_kind":"pith_short_16","alias_value":"5UXZVVEIIZMPOJCJ","created_at":"2026-08-04T00:33:03Z"},{"alias_kind":"pith_short_8","alias_value":"5UXZVVEI","created_at":"2026-08-04T00:33:03Z"}],"graph_snapshots":[{"event_id":"sha256:408b60e471a273bc0318e5f95816b5076cf3092c0edf1042c8f58f849179295e","target":"graph","created_at":"2026-08-04T00:33:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2608.00144/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Membership inference (MIA) on language models is usually summarised by an aggregate ROC-AUC, but such evaluations are confounded: model-free blind baselines separate members from non-members from surface text alone. We study black-box, sampling-based training-data leakage through a probabilistic lens, treating N samples from p(.|x) as an estimate of the output distribution and casting leakage signals as functionals of it. We extend the blind-baseline critique into the sampling regime: on WikiMIA a blind bag-of-words classifier reaches AUC 0.97 (TPR 0.90 at 5% FPR) and sampling adds nothing, wh","authors_text":"Victor Maricato","cross_cats":["cs.CL","cs.CR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-31T15:44:43Z","title":"Leak It: A Probabilistic Approach to Training-Data Extraction from Black-Box Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00144","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e983b05b13b77c0f0c9af8a623fd3e277b915fa466f9011cc323f1bed407f80f","target":"record","created_at":"2026-08-04T00:33:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d72cc5b3fece72928462c260e5d5284ab71ce0d2017b0b84f49c7118a17a6318","cross_cats_sorted":["cs.CL","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-31T15:44:43Z","title_canon_sha256":"cf9def1fb7be33597019208bbe14514be554560c1ed944f450deef9c5a77ec71"},"schema_version":"1.0","source":{"id":"2608.00144","kind":"arxiv","version":1}},"canonical_sha256":"ed2f9ad4884658f72449c16805bbd997ff24cc9958c0d58b35721f8f0cc39306","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ed2f9ad4884658f72449c16805bbd997ff24cc9958c0d58b35721f8f0cc39306","first_computed_at":"2026-08-04T00:33:03.137082Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-04T00:33:03.137082Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"y//Y0kAPxkxS21pJhqJl6M3VZO6Yb+lgSrkZUHbGKVB4HvF1tb7aj3r831j2t9zFogbBYinLhBIxXZ8RkKhTAg==","signature_status":"signed_v1","signed_at":"2026-08-04T00:33:03.138477Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.00144","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e983b05b13b77c0f0c9af8a623fd3e277b915fa466f9011cc323f1bed407f80f","sha256:408b60e471a273bc0318e5f95816b5076cf3092c0edf1042c8f58f849179295e"],"state_sha256":"8e1e94942254c8ddfe0cb1d3ed6cd84be775d617f033ca0c495b368163bf4ee0"}