{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VUOZONLPLXPSIZIFWXTNCJOJ4U","short_pith_number":"pith:VUOZONLP","schema_version":"1.0","canonical_sha256":"ad1d97356f5ddf246505b5e6d125c9e51a3d751f0b43b3b78b0a00692841aeae","source":{"kind":"arxiv","id":"2502.20629","version":1},"attestation_state":"computed","paper":{"title":"Towards Privacy-Preserving Split Learning: Destabilizing Adversarial Inference and Reconstruction Attacks in the Cloud","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Ali Ghorbani, Amir David, Griffin Higgins, Roozbeh Razavi-Far, Tongyu Ge, Xichen Zhang","submitted_at":"2025-02-28T01:24:33Z","abstract_excerpt":"This work aims to provide both privacy and utility within a split learning framework while considering both forward attribute inference and backward reconstruction attacks. To address this, a novel approach has been proposed, which makes use of class activation maps and autoencoders as a plug-in strategy aiming to increase the user's privacy and destabilize an adversary. The proposed approach is compared with a dimensionality-reduction-based plug-in strategy, which makes use of principal component analysis to transform the feature map onto a lower-dimensional feature space. Our work shows 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":"2502.20629","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-28T01:24:33Z","cross_cats_sorted":[],"title_canon_sha256":"b5a409daea32de895e55590de7f4c54152932a38f9342bcd3054925ae8673f83","abstract_canon_sha256":"1502e5e86da7680d5b1a57848dff2a0933d97d1de2807f6b44ffc7cfb482e48b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:36.380955Z","signature_b64":"eiIKH76yySXp/5Z+4pPACPX8Rl7fg7ebuj1Qtwmm+rPS3hA2IMKfTDE1BF77+t68Zkv+8ZFSOPAETeM9NsP6CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad1d97356f5ddf246505b5e6d125c9e51a3d751f0b43b3b78b0a00692841aeae","last_reissued_at":"2026-07-05T10:21:36.380446Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:36.380446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Privacy-Preserving Split Learning: Destabilizing Adversarial Inference and Reconstruction Attacks in the Cloud","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Ali Ghorbani, Amir David, Griffin Higgins, Roozbeh Razavi-Far, Tongyu Ge, Xichen Zhang","submitted_at":"2025-02-28T01:24:33Z","abstract_excerpt":"This work aims to provide both privacy and utility within a split learning framework while considering both forward attribute inference and backward reconstruction attacks. To address this, a novel approach has been proposed, which makes use of class activation maps and autoencoders as a plug-in strategy aiming to increase the user's privacy and destabilize an adversary. The proposed approach is compared with a dimensionality-reduction-based plug-in strategy, which makes use of principal component analysis to transform the feature map onto a lower-dimensional feature space. Our work shows that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20629","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/2502.20629/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":"2502.20629","created_at":"2026-07-05T10:21:36.380502+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.20629v1","created_at":"2026-07-05T10:21:36.380502+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20629","created_at":"2026-07-05T10:21:36.380502+00:00"},{"alias_kind":"pith_short_12","alias_value":"VUOZONLPLXPS","created_at":"2026-07-05T10:21:36.380502+00:00"},{"alias_kind":"pith_short_16","alias_value":"VUOZONLPLXPSIZIF","created_at":"2026-07-05T10:21:36.380502+00:00"},{"alias_kind":"pith_short_8","alias_value":"VUOZONLP","created_at":"2026-07-05T10:21:36.380502+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/VUOZONLPLXPSIZIFWXTNCJOJ4U","json":"https://pith.science/pith/VUOZONLPLXPSIZIFWXTNCJOJ4U.json","graph_json":"https://pith.science/api/pith-number/VUOZONLPLXPSIZIFWXTNCJOJ4U/graph.json","events_json":"https://pith.science/api/pith-number/VUOZONLPLXPSIZIFWXTNCJOJ4U/events.json","paper":"https://pith.science/paper/VUOZONLP"},"agent_actions":{"view_html":"https://pith.science/pith/VUOZONLPLXPSIZIFWXTNCJOJ4U","download_json":"https://pith.science/pith/VUOZONLPLXPSIZIFWXTNCJOJ4U.json","view_paper":"https://pith.science/paper/VUOZONLP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.20629&json=true","fetch_graph":"https://pith.science/api/pith-number/VUOZONLPLXPSIZIFWXTNCJOJ4U/graph.json","fetch_events":"https://pith.science/api/pith-number/VUOZONLPLXPSIZIFWXTNCJOJ4U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VUOZONLPLXPSIZIFWXTNCJOJ4U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VUOZONLPLXPSIZIFWXTNCJOJ4U/action/storage_attestation","attest_author":"https://pith.science/pith/VUOZONLPLXPSIZIFWXTNCJOJ4U/action/author_attestation","sign_citation":"https://pith.science/pith/VUOZONLPLXPSIZIFWXTNCJOJ4U/action/citation_signature","submit_replication":"https://pith.science/pith/VUOZONLPLXPSIZIFWXTNCJOJ4U/action/replication_record"}},"created_at":"2026-07-05T10:21:36.380502+00:00","updated_at":"2026-07-05T10:21:36.380502+00:00"}