{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:NJBCFQRIVDVV7IJBZ3ADXJXX66","short_pith_number":"pith:NJBCFQRI","schema_version":"1.0","canonical_sha256":"6a4222c228a8eb5fa121cec03ba6f7f7ad844b37d3211c2aa6c92ffa386fbace","source":{"kind":"arxiv","id":"1812.00292","version":4},"attestation_state":"computed","paper":{"title":"SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Edward Chou, Florian Tram\\`er, Giancarlo Pellegrino","submitted_at":"2018-12-02T00:03:07Z","abstract_excerpt":"SentiNet is a novel detection framework for localized universal attacks on neural networks. These attacks restrict adversarial noise to contiguous portions of an image and are reusable with different images -- constraints that prove useful for generating physically-realizable attacks. Unlike most other works on adversarial detection, SentiNet does not require training a model or preknowledge of an attack prior to detection. Our approach is appealing due to the large number of possible mechanisms and attack-vectors that an attack-specific defense would have to consider. By leveraging the neural"},"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":"1812.00292","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2018-12-02T00:03:07Z","cross_cats_sorted":[],"title_canon_sha256":"68753de6545d446df61017caf445d57d31a6c8d8bf7a2162f165b0ffacd60d70","abstract_canon_sha256":"495687b435298d7d520f0042eb65ae4cf7a191de7838156f576599c4c8d2838b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:01:33.624899Z","signature_b64":"quKU22uO2hYW0KND1lBTjgKidN2jpe/YaKnXjEDfVgxCjRSFCKFA+QjTz+HqSnD8QpjtjZUXwk6ek79pXGiGCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a4222c228a8eb5fa121cec03ba6f7f7ad844b37d3211c2aa6c92ffa386fbace","last_reissued_at":"2026-07-05T01:01:33.624270Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:01:33.624270Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Edward Chou, Florian Tram\\`er, Giancarlo Pellegrino","submitted_at":"2018-12-02T00:03:07Z","abstract_excerpt":"SentiNet is a novel detection framework for localized universal attacks on neural networks. These attacks restrict adversarial noise to contiguous portions of an image and are reusable with different images -- constraints that prove useful for generating physically-realizable attacks. Unlike most other works on adversarial detection, SentiNet does not require training a model or preknowledge of an attack prior to detection. Our approach is appealing due to the large number of possible mechanisms and attack-vectors that an attack-specific defense would have to consider. By leveraging the neural"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.00292","kind":"arxiv","version":4},"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/1812.00292/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":"1812.00292","created_at":"2026-07-05T01:01:33.624362+00:00"},{"alias_kind":"arxiv_version","alias_value":"1812.00292v4","created_at":"2026-07-05T01:01:33.624362+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.00292","created_at":"2026-07-05T01:01:33.624362+00:00"},{"alias_kind":"pith_short_12","alias_value":"NJBCFQRIVDVV","created_at":"2026-07-05T01:01:33.624362+00:00"},{"alias_kind":"pith_short_16","alias_value":"NJBCFQRIVDVV7IJB","created_at":"2026-07-05T01:01:33.624362+00:00"},{"alias_kind":"pith_short_8","alias_value":"NJBCFQRI","created_at":"2026-07-05T01:01:33.624362+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.05224","citing_title":"A Survey on Backdoor Threats in Large Language Models (LLMs): Attacks, Defenses, and Evaluations","ref_index":220,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66","json":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66.json","graph_json":"https://pith.science/api/pith-number/NJBCFQRIVDVV7IJBZ3ADXJXX66/graph.json","events_json":"https://pith.science/api/pith-number/NJBCFQRIVDVV7IJBZ3ADXJXX66/events.json","paper":"https://pith.science/paper/NJBCFQRI"},"agent_actions":{"view_html":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66","download_json":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66.json","view_paper":"https://pith.science/paper/NJBCFQRI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1812.00292&json=true","fetch_graph":"https://pith.science/api/pith-number/NJBCFQRIVDVV7IJBZ3ADXJXX66/graph.json","fetch_events":"https://pith.science/api/pith-number/NJBCFQRIVDVV7IJBZ3ADXJXX66/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66/action/storage_attestation","attest_author":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66/action/author_attestation","sign_citation":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66/action/citation_signature","submit_replication":"https://pith.science/pith/NJBCFQRIVDVV7IJBZ3ADXJXX66/action/replication_record"}},"created_at":"2026-07-05T01:01:33.624362+00:00","updated_at":"2026-07-05T01:01:33.624362+00:00"}