{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:VVSRQPCYY5ND2ALDFF7KYCBOHI","short_pith_number":"pith:VVSRQPCY","schema_version":"1.0","canonical_sha256":"ad65183c58c75a3d0163297eac082e3a38a550479fff3fe198acc67892d48792","source":{"kind":"arxiv","id":"2310.08259","version":1},"attestation_state":"computed","paper":{"title":"Invisible Threats: Backdoor Attack in OCR Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"cs.CR","authors_text":"Luca Pajola, Mauro Conti, Nicola Farronato, Stefanos Koffas, Stjepan Picek","submitted_at":"2023-10-12T12:05:51Z","abstract_excerpt":"Optical Character Recognition (OCR) is a widely used tool to extract text from scanned documents. Today, the state-of-the-art is achieved by exploiting deep neural networks. However, the cost of this performance is paid at the price of system vulnerability. For instance, in backdoor attacks, attackers compromise the training phase by inserting a backdoor in the victim's model that will be activated at testing time by specific patterns while leaving the overall model performance intact. This work proposes a backdoor attack for OCR resulting in the injection of non-readable characters from malic"},"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":"2310.08259","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2023-10-12T12:05:51Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"ad6561febe1bb1694c392843cc3935c872e647bf925bba088ec87e1c632d0e96","abstract_canon_sha256":"87aad5200b81c57adda41f963eccd5d4cd01947e05fc8cb2002b9e517e7abefb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:00:13.078446Z","signature_b64":"ogUN4VEAH6WkmKD0WRa0YBNdWhnjGjsORXoC8+SCc8DHBNwLz4XHcEamX8dUJl7rgRS7nZHFvUq0pghbTK5RBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad65183c58c75a3d0163297eac082e3a38a550479fff3fe198acc67892d48792","last_reissued_at":"2026-07-05T07:00:13.078019Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:00:13.078019Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Invisible Threats: Backdoor Attack in OCR Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"cs.CR","authors_text":"Luca Pajola, Mauro Conti, Nicola Farronato, Stefanos Koffas, Stjepan Picek","submitted_at":"2023-10-12T12:05:51Z","abstract_excerpt":"Optical Character Recognition (OCR) is a widely used tool to extract text from scanned documents. Today, the state-of-the-art is achieved by exploiting deep neural networks. However, the cost of this performance is paid at the price of system vulnerability. For instance, in backdoor attacks, attackers compromise the training phase by inserting a backdoor in the victim's model that will be activated at testing time by specific patterns while leaving the overall model performance intact. This work proposes a backdoor attack for OCR resulting in the injection of non-readable characters from malic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.08259","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/2310.08259/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":"2310.08259","created_at":"2026-07-05T07:00:13.078075+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.08259v1","created_at":"2026-07-05T07:00:13.078075+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.08259","created_at":"2026-07-05T07:00:13.078075+00:00"},{"alias_kind":"pith_short_12","alias_value":"VVSRQPCYY5ND","created_at":"2026-07-05T07:00:13.078075+00:00"},{"alias_kind":"pith_short_16","alias_value":"VVSRQPCYY5ND2ALD","created_at":"2026-07-05T07:00:13.078075+00:00"},{"alias_kind":"pith_short_8","alias_value":"VVSRQPCY","created_at":"2026-07-05T07:00:13.078075+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.03993","citing_title":"LaserGuider: A Laser Based Physical Backdoor Attack against Deep Neural Networks","ref_index":9,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI","json":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI.json","graph_json":"https://pith.science/api/pith-number/VVSRQPCYY5ND2ALDFF7KYCBOHI/graph.json","events_json":"https://pith.science/api/pith-number/VVSRQPCYY5ND2ALDFF7KYCBOHI/events.json","paper":"https://pith.science/paper/VVSRQPCY"},"agent_actions":{"view_html":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI","download_json":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI.json","view_paper":"https://pith.science/paper/VVSRQPCY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.08259&json=true","fetch_graph":"https://pith.science/api/pith-number/VVSRQPCYY5ND2ALDFF7KYCBOHI/graph.json","fetch_events":"https://pith.science/api/pith-number/VVSRQPCYY5ND2ALDFF7KYCBOHI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI/action/storage_attestation","attest_author":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI/action/author_attestation","sign_citation":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI/action/citation_signature","submit_replication":"https://pith.science/pith/VVSRQPCYY5ND2ALDFF7KYCBOHI/action/replication_record"}},"created_at":"2026-07-05T07:00:13.078075+00:00","updated_at":"2026-07-05T07:00:13.078075+00:00"}