{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4PJIIXASNFXKSYCDML7FKJPMXH","short_pith_number":"pith:4PJIIXAS","schema_version":"1.0","canonical_sha256":"e3d2845c12696ea9604362fe5525ecb9d0ec169decec15c5df093169e0ddee2a","source":{"kind":"arxiv","id":"2607.18342","version":1},"attestation_state":"computed","paper":{"title":"PRISM: Sensitivity-Aware PolynoMial PRuning for EffIcient Neural Network Encryption","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.DC","cs.LG"],"primary_cat":"cs.CR","authors_text":"Mahdi Taheri, Sahaj Majavdia","submitted_at":"2026-07-20T05:47:06Z","abstract_excerpt":"Structured pruning is essential for making neural network inference feasible under homomorphic encryption (HE), yet its impact on model reliability has remained unexplored. This paper presents a systematic reliability characterization of pruned CKKS-encrypted neural networks and introduces Polynomial-Sensitivity-Aware Pruning (PSAP), a structured pruning method that is inherently reliability-aware. PSAP scores filters jointly by weight magnitude, polynomial activation sensitivity, and rotation cost, which concentrates pruning in fault-tolerant regions. Across two architectures, two datasets, t"},"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":"2607.18342","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-20T05:47:06Z","cross_cats_sorted":["cs.AI","cs.AR","cs.DC","cs.LG"],"title_canon_sha256":"ed0f7222116b8fbf4346085f2d95c95f8b820ff726be580fda0dd10a800a3aa2","abstract_canon_sha256":"74c94ac1d21b352b308b1a62512a44305947f150af8012ba57cf9695d117b5a0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-22T00:22:42.584254Z","signature_b64":"HUxnqSfa8pQLicAydQlHO1o1Vju9GtPBrf6IqljDx/IY2z0sG5Cg2j6PoO38mpeTKO6yb6q3BmmPNA1eVE4gDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3d2845c12696ea9604362fe5525ecb9d0ec169decec15c5df093169e0ddee2a","last_reissued_at":"2026-07-22T00:22:42.583383Z","signature_status":"signed_v1","first_computed_at":"2026-07-22T00:22:42.583383Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PRISM: Sensitivity-Aware PolynoMial PRuning for EffIcient Neural Network Encryption","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.DC","cs.LG"],"primary_cat":"cs.CR","authors_text":"Mahdi Taheri, Sahaj Majavdia","submitted_at":"2026-07-20T05:47:06Z","abstract_excerpt":"Structured pruning is essential for making neural network inference feasible under homomorphic encryption (HE), yet its impact on model reliability has remained unexplored. This paper presents a systematic reliability characterization of pruned CKKS-encrypted neural networks and introduces Polynomial-Sensitivity-Aware Pruning (PSAP), a structured pruning method that is inherently reliability-aware. PSAP scores filters jointly by weight magnitude, polynomial activation sensitivity, and rotation cost, which concentrates pruning in fault-tolerant regions. Across two architectures, two datasets, t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.18342","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/2607.18342/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":"2607.18342","created_at":"2026-07-22T00:22:42.583789+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.18342v1","created_at":"2026-07-22T00:22:42.583789+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.18342","created_at":"2026-07-22T00:22:42.583789+00:00"},{"alias_kind":"pith_short_12","alias_value":"4PJIIXASNFXK","created_at":"2026-07-22T00:22:42.583789+00:00"},{"alias_kind":"pith_short_16","alias_value":"4PJIIXASNFXKSYCD","created_at":"2026-07-22T00:22:42.583789+00:00"},{"alias_kind":"pith_short_8","alias_value":"4PJIIXAS","created_at":"2026-07-22T00:22:42.583789+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/4PJIIXASNFXKSYCDML7FKJPMXH","json":"https://pith.science/pith/4PJIIXASNFXKSYCDML7FKJPMXH.json","graph_json":"https://pith.science/api/pith-number/4PJIIXASNFXKSYCDML7FKJPMXH/graph.json","events_json":"https://pith.science/api/pith-number/4PJIIXASNFXKSYCDML7FKJPMXH/events.json","paper":"https://pith.science/paper/4PJIIXAS"},"agent_actions":{"view_html":"https://pith.science/pith/4PJIIXASNFXKSYCDML7FKJPMXH","download_json":"https://pith.science/pith/4PJIIXASNFXKSYCDML7FKJPMXH.json","view_paper":"https://pith.science/paper/4PJIIXAS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.18342&json=true","fetch_graph":"https://pith.science/api/pith-number/4PJIIXASNFXKSYCDML7FKJPMXH/graph.json","fetch_events":"https://pith.science/api/pith-number/4PJIIXASNFXKSYCDML7FKJPMXH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4PJIIXASNFXKSYCDML7FKJPMXH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4PJIIXASNFXKSYCDML7FKJPMXH/action/storage_attestation","attest_author":"https://pith.science/pith/4PJIIXASNFXKSYCDML7FKJPMXH/action/author_attestation","sign_citation":"https://pith.science/pith/4PJIIXASNFXKSYCDML7FKJPMXH/action/citation_signature","submit_replication":"https://pith.science/pith/4PJIIXASNFXKSYCDML7FKJPMXH/action/replication_record"}},"created_at":"2026-07-22T00:22:42.583789+00:00","updated_at":"2026-07-22T00:22:42.583789+00:00"}