{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:L3Q6CLFEN7D2IIRVODQHKNBLFF","short_pith_number":"pith:L3Q6CLFE","schema_version":"1.0","canonical_sha256":"5ee1e12ca46fc7a4223570e075342b2958061bfc49734ffa827ac66404ae6c1c","source":{"kind":"arxiv","id":"2606.26021","version":1},"attestation_state":"computed","paper":{"title":"Privacy Vulnerabilities of Attention Layers in Tabular Foundation Models and Protection of High-Risk Queries","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Maxime Cordy, T\\^ania Carvalho","submitted_at":"2026-06-24T16:42:21Z","abstract_excerpt":"Tabular foundation models are commonly assumed to present limited privacy concerns as they are often pre-trained on large collections of synthetic data. However, these models leverage in-context learning, where sensitive records may be provided directly at inference time as labelled context examples. In this paper, we demonstrate that predictions generated via the attention mechanism leak sufficient information to enable effective Membership Inference Attacks (MIAs). To highlight this vulnerability, we propose AMIA (Attention-based Membership Inference Attack), a shadow-model-free attack that "},"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":"2606.26021","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-24T16:42:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"41f07b1185f8fe0a9be1de1850ae442a683b1a5812543d0a3928120ecdc4216c","abstract_canon_sha256":"8400e75fb3eeee31f95757eef2129a4991aab14608aed2f78d8d6bc28761420a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-25T01:18:46.081767Z","signature_b64":"jr9ZgniRD2v8LeZaJ58MOpjy149m7ePG6C+gPtCIFUM+io5/no03iv3BpQN6pHQvWlHH+hadajQquTpVLwxSBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5ee1e12ca46fc7a4223570e075342b2958061bfc49734ffa827ac66404ae6c1c","last_reissued_at":"2026-06-25T01:18:46.081383Z","signature_status":"signed_v1","first_computed_at":"2026-06-25T01:18:46.081383Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Privacy Vulnerabilities of Attention Layers in Tabular Foundation Models and Protection of High-Risk Queries","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Maxime Cordy, T\\^ania Carvalho","submitted_at":"2026-06-24T16:42:21Z","abstract_excerpt":"Tabular foundation models are commonly assumed to present limited privacy concerns as they are often pre-trained on large collections of synthetic data. However, these models leverage in-context learning, where sensitive records may be provided directly at inference time as labelled context examples. In this paper, we demonstrate that predictions generated via the attention mechanism leak sufficient information to enable effective Membership Inference Attacks (MIAs). To highlight this vulnerability, we propose AMIA (Attention-based Membership Inference Attack), a shadow-model-free attack that "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26021","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/2606.26021/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":"2606.26021","created_at":"2026-06-25T01:18:46.081446+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26021v1","created_at":"2026-06-25T01:18:46.081446+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26021","created_at":"2026-06-25T01:18:46.081446+00:00"},{"alias_kind":"pith_short_12","alias_value":"L3Q6CLFEN7D2","created_at":"2026-06-25T01:18:46.081446+00:00"},{"alias_kind":"pith_short_16","alias_value":"L3Q6CLFEN7D2IIRV","created_at":"2026-06-25T01:18:46.081446+00:00"},{"alias_kind":"pith_short_8","alias_value":"L3Q6CLFE","created_at":"2026-06-25T01:18:46.081446+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/L3Q6CLFEN7D2IIRVODQHKNBLFF","json":"https://pith.science/pith/L3Q6CLFEN7D2IIRVODQHKNBLFF.json","graph_json":"https://pith.science/api/pith-number/L3Q6CLFEN7D2IIRVODQHKNBLFF/graph.json","events_json":"https://pith.science/api/pith-number/L3Q6CLFEN7D2IIRVODQHKNBLFF/events.json","paper":"https://pith.science/paper/L3Q6CLFE"},"agent_actions":{"view_html":"https://pith.science/pith/L3Q6CLFEN7D2IIRVODQHKNBLFF","download_json":"https://pith.science/pith/L3Q6CLFEN7D2IIRVODQHKNBLFF.json","view_paper":"https://pith.science/paper/L3Q6CLFE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26021&json=true","fetch_graph":"https://pith.science/api/pith-number/L3Q6CLFEN7D2IIRVODQHKNBLFF/graph.json","fetch_events":"https://pith.science/api/pith-number/L3Q6CLFEN7D2IIRVODQHKNBLFF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L3Q6CLFEN7D2IIRVODQHKNBLFF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L3Q6CLFEN7D2IIRVODQHKNBLFF/action/storage_attestation","attest_author":"https://pith.science/pith/L3Q6CLFEN7D2IIRVODQHKNBLFF/action/author_attestation","sign_citation":"https://pith.science/pith/L3Q6CLFEN7D2IIRVODQHKNBLFF/action/citation_signature","submit_replication":"https://pith.science/pith/L3Q6CLFEN7D2IIRVODQHKNBLFF/action/replication_record"}},"created_at":"2026-06-25T01:18:46.081446+00:00","updated_at":"2026-06-25T01:18:46.081446+00:00"}