{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:PHIGMRJK4XWNOXSMVKMPVZYBKU","short_pith_number":"pith:PHIGMRJK","schema_version":"1.0","canonical_sha256":"79d066452ae5ecd75e4caa98fae7015528b3a816ce2923abaab3b203471349d3","source":{"kind":"arxiv","id":"2002.08313","version":2},"attestation_state":"computed","paper":{"title":"NNoculation: Catching BadNets in the Wild","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Akshaj Kumar Veldanda, Benjamin Tan, Brendan Dolan-Gavitt, Farshad Khorrami, Kang Liu, Prashanth Krishnamurthy, Ramesh Karri, Siddharth Garg","submitted_at":"2020-02-19T17:51:21Z","abstract_excerpt":"This paper proposes a novel two-stage defense (NNoculation) against backdoored neural networks (BadNets) that, repairs a BadNet both pre-deployment and online in response to backdoored test inputs encountered in the field. In the pre-deployment stage, NNoculation retrains the BadNet with random perturbations of clean validation inputs to partially reduce the adversarial impact of a backdoor. Post-deployment, NNoculation detects and quarantines backdoored test inputs by recording disagreements between the original and pre-deployment patched networks. A CycleGAN is then trained to learn transfor"},"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":"2002.08313","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2020-02-19T17:51:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"de259671f976d7e97393b2dd0853d7784e838a9a5c3fcd7393e3ecef8b056a19","abstract_canon_sha256":"eb73ab2cd3da78f416a1a6a70a42e2a9bc9ed851e1704adcaf2d1b24053518d1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:31:21.116739Z","signature_b64":"+g/GyaURDiMLnOvoQeMKNglI8Z7r7ycRPjjowoABSKCQRDn8EL2CddwpRXufq+arPB+mADkXSsdMFZzUKfbCCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"79d066452ae5ecd75e4caa98fae7015528b3a816ce2923abaab3b203471349d3","last_reissued_at":"2026-07-05T03:31:21.116161Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:31:21.116161Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NNoculation: Catching BadNets in the Wild","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Akshaj Kumar Veldanda, Benjamin Tan, Brendan Dolan-Gavitt, Farshad Khorrami, Kang Liu, Prashanth Krishnamurthy, Ramesh Karri, Siddharth Garg","submitted_at":"2020-02-19T17:51:21Z","abstract_excerpt":"This paper proposes a novel two-stage defense (NNoculation) against backdoored neural networks (BadNets) that, repairs a BadNet both pre-deployment and online in response to backdoored test inputs encountered in the field. In the pre-deployment stage, NNoculation retrains the BadNet with random perturbations of clean validation inputs to partially reduce the adversarial impact of a backdoor. Post-deployment, NNoculation detects and quarantines backdoored test inputs by recording disagreements between the original and pre-deployment patched networks. A CycleGAN is then trained to learn transfor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08313","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/2002.08313/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":"2002.08313","created_at":"2026-07-05T03:31:21.116226+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.08313v2","created_at":"2026-07-05T03:31:21.116226+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08313","created_at":"2026-07-05T03:31:21.116226+00:00"},{"alias_kind":"pith_short_12","alias_value":"PHIGMRJK4XWN","created_at":"2026-07-05T03:31:21.116226+00:00"},{"alias_kind":"pith_short_16","alias_value":"PHIGMRJK4XWNOXSM","created_at":"2026-07-05T03:31:21.116226+00:00"},{"alias_kind":"pith_short_8","alias_value":"PHIGMRJK","created_at":"2026-07-05T03:31:21.116226+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.13864","citing_title":"Inevitable Encounters: Backdoor Attacks Involving Lossy Compression","ref_index":65,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU","json":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU.json","graph_json":"https://pith.science/api/pith-number/PHIGMRJK4XWNOXSMVKMPVZYBKU/graph.json","events_json":"https://pith.science/api/pith-number/PHIGMRJK4XWNOXSMVKMPVZYBKU/events.json","paper":"https://pith.science/paper/PHIGMRJK"},"agent_actions":{"view_html":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU","download_json":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU.json","view_paper":"https://pith.science/paper/PHIGMRJK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.08313&json=true","fetch_graph":"https://pith.science/api/pith-number/PHIGMRJK4XWNOXSMVKMPVZYBKU/graph.json","fetch_events":"https://pith.science/api/pith-number/PHIGMRJK4XWNOXSMVKMPVZYBKU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU/action/storage_attestation","attest_author":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU/action/author_attestation","sign_citation":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU/action/citation_signature","submit_replication":"https://pith.science/pith/PHIGMRJK4XWNOXSMVKMPVZYBKU/action/replication_record"}},"created_at":"2026-07-05T03:31:21.116226+00:00","updated_at":"2026-07-05T03:31:21.116226+00:00"}