{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XBUUH2RCAIDH2POWZ5F5G2ZCAO","short_pith_number":"pith:XBUUH2RC","schema_version":"1.0","canonical_sha256":"b86943ea2202067d3dd6cf4bd36b2203822222f798703436afff5a270ec3ee16","source":{"kind":"arxiv","id":"2504.12806","version":2},"attestation_state":"computed","paper":{"title":"A Numerical Gradient Inversion Attack in Variational Quantum Neural-Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Chun-Fu Chen, Georgios Papadopoulos, Jamie Heredge, Marco Pistoia, Niraj Kumar, Shaltiel Eloul, Yash Satsangi","submitted_at":"2025-04-17T10:12:38Z","abstract_excerpt":"The loss landscape of Variational Quantum Neural Networks (VQNNs) is characterized by local minima that grow exponentially with increasing qubits. Because of this, it is more challenging to recover information from model gradients during training compared to classical Neural Networks (NNs). In this paper we present a numerical scheme that successfully reconstructs input training, real-world, practical data from trainable VQNNs' gradients. Our scheme is based on gradient inversion that works by combining gradients estimation with the finite difference method and adaptive low-pass filtering. The"},"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":"2504.12806","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-17T10:12:38Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"4e95abd9eef2204cad97ec66df407cbf4ba4c603f60f5f5802d00f85c5d93918","abstract_canon_sha256":"b6f07cf332733a544873fd17bb01f5d2b08592c92d36fdfc0648e9a2432de210"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:59:28.927588Z","signature_b64":"h5m/+jaCOuGQu4ae8EKufS166NoFG63NF7Ye9i0bHzYdjvYGQeNdFuKL9eZxt6alcz+WhmwLjpzwK8wHMM+TAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b86943ea2202067d3dd6cf4bd36b2203822222f798703436afff5a270ec3ee16","last_reissued_at":"2026-07-05T10:59:28.927082Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:59:28.927082Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Numerical Gradient Inversion Attack in Variational Quantum Neural-Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Chun-Fu Chen, Georgios Papadopoulos, Jamie Heredge, Marco Pistoia, Niraj Kumar, Shaltiel Eloul, Yash Satsangi","submitted_at":"2025-04-17T10:12:38Z","abstract_excerpt":"The loss landscape of Variational Quantum Neural Networks (VQNNs) is characterized by local minima that grow exponentially with increasing qubits. Because of this, it is more challenging to recover information from model gradients during training compared to classical Neural Networks (NNs). In this paper we present a numerical scheme that successfully reconstructs input training, real-world, practical data from trainable VQNNs' gradients. Our scheme is based on gradient inversion that works by combining gradients estimation with the finite difference method and adaptive low-pass filtering. The"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12806","kind":"arxiv","version":2},"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/2504.12806/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":"2504.12806","created_at":"2026-07-05T10:59:28.927140+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.12806v2","created_at":"2026-07-05T10:59:28.927140+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12806","created_at":"2026-07-05T10:59:28.927140+00:00"},{"alias_kind":"pith_short_12","alias_value":"XBUUH2RCAIDH","created_at":"2026-07-05T10:59:28.927140+00:00"},{"alias_kind":"pith_short_16","alias_value":"XBUUH2RCAIDH2POW","created_at":"2026-07-05T10:59:28.927140+00:00"},{"alias_kind":"pith_short_8","alias_value":"XBUUH2RC","created_at":"2026-07-05T10:59:28.927140+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.18821","citing_title":"Quantum Adversarial Machine Learning: From Classical Adaptations to Quantum-Native Methods","ref_index":65,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO","json":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO.json","graph_json":"https://pith.science/api/pith-number/XBUUH2RCAIDH2POWZ5F5G2ZCAO/graph.json","events_json":"https://pith.science/api/pith-number/XBUUH2RCAIDH2POWZ5F5G2ZCAO/events.json","paper":"https://pith.science/paper/XBUUH2RC"},"agent_actions":{"view_html":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO","download_json":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO.json","view_paper":"https://pith.science/paper/XBUUH2RC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.12806&json=true","fetch_graph":"https://pith.science/api/pith-number/XBUUH2RCAIDH2POWZ5F5G2ZCAO/graph.json","fetch_events":"https://pith.science/api/pith-number/XBUUH2RCAIDH2POWZ5F5G2ZCAO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO/action/storage_attestation","attest_author":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO/action/author_attestation","sign_citation":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO/action/citation_signature","submit_replication":"https://pith.science/pith/XBUUH2RCAIDH2POWZ5F5G2ZCAO/action/replication_record"}},"created_at":"2026-07-05T10:59:28.927140+00:00","updated_at":"2026-07-05T10:59:28.927140+00:00"}