{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IOAD2JSHV4QRRA7TOB26RIOOIK","short_pith_number":"pith:IOAD2JSH","schema_version":"1.0","canonical_sha256":"43803d2647af211883f37075e8a1ce42a1a8c38eec3c7ed8d7d717901b95125e","source":{"kind":"arxiv","id":"2410.21146","version":1},"attestation_state":"computed","paper":{"title":"Palisade -- Prompt Injection Detection Framework","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Aashishkumar, G Divya, Nandavardhan R, Sahasra Kokkula, Somanathan R","submitted_at":"2024-10-28T15:47:03Z","abstract_excerpt":"The advent of Large Language Models LLMs marks a milestone in Artificial Intelligence, altering how machines comprehend and generate human language. However, LLMs are vulnerable to malicious prompt injection attacks, where crafted inputs manipulate the models behavior in unintended ways, compromising system integrity and causing incorrect outcomes. Conventional detection methods rely on static, rule-based approaches, which often fail against sophisticated threats like abnormal token sequences and alias substitutions, leading to limited adaptability and higher rates of false positives and false"},"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":"2410.21146","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-28T15:47:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"83f709ce920ef105004ba078b37891a539699ef53528f3211fb8f3103c9e725e","abstract_canon_sha256":"f474bd35eb2eb1a717c018fd76f4378503fd433aa9ffd7a74700e6900b9f5691"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:15.980583Z","signature_b64":"qReUYSGku9UFr8NVKdPZTARTO2Rm+Qbl4wgHFyVgbUhyaI6GNBIWnWfk0xpGRbA85u79m/mI5MRQHywP32PVAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43803d2647af211883f37075e8a1ce42a1a8c38eec3c7ed8d7d717901b95125e","last_reissued_at":"2026-07-05T09:27:15.980085Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:15.980085Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Palisade -- Prompt Injection Detection Framework","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Aashishkumar, G Divya, Nandavardhan R, Sahasra Kokkula, Somanathan R","submitted_at":"2024-10-28T15:47:03Z","abstract_excerpt":"The advent of Large Language Models LLMs marks a milestone in Artificial Intelligence, altering how machines comprehend and generate human language. However, LLMs are vulnerable to malicious prompt injection attacks, where crafted inputs manipulate the models behavior in unintended ways, compromising system integrity and causing incorrect outcomes. Conventional detection methods rely on static, rule-based approaches, which often fail against sophisticated threats like abnormal token sequences and alias substitutions, leading to limited adaptability and higher rates of false positives and false"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.21146","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/2410.21146/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":"2410.21146","created_at":"2026-07-05T09:27:15.980146+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.21146v1","created_at":"2026-07-05T09:27:15.980146+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.21146","created_at":"2026-07-05T09:27:15.980146+00:00"},{"alias_kind":"pith_short_12","alias_value":"IOAD2JSHV4QR","created_at":"2026-07-05T09:27:15.980146+00:00"},{"alias_kind":"pith_short_16","alias_value":"IOAD2JSHV4QRRA7T","created_at":"2026-07-05T09:27:15.980146+00:00"},{"alias_kind":"pith_short_8","alias_value":"IOAD2JSH","created_at":"2026-07-05T09:27:15.980146+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.16682","citing_title":"The Task Shield: Enforcing Task Alignment to Defend Against Indirect Prompt Injection in LLM Agents","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK","json":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK.json","graph_json":"https://pith.science/api/pith-number/IOAD2JSHV4QRRA7TOB26RIOOIK/graph.json","events_json":"https://pith.science/api/pith-number/IOAD2JSHV4QRRA7TOB26RIOOIK/events.json","paper":"https://pith.science/paper/IOAD2JSH"},"agent_actions":{"view_html":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK","download_json":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK.json","view_paper":"https://pith.science/paper/IOAD2JSH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.21146&json=true","fetch_graph":"https://pith.science/api/pith-number/IOAD2JSHV4QRRA7TOB26RIOOIK/graph.json","fetch_events":"https://pith.science/api/pith-number/IOAD2JSHV4QRRA7TOB26RIOOIK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK/action/storage_attestation","attest_author":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK/action/author_attestation","sign_citation":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK/action/citation_signature","submit_replication":"https://pith.science/pith/IOAD2JSHV4QRRA7TOB26RIOOIK/action/replication_record"}},"created_at":"2026-07-05T09:27:15.980146+00:00","updated_at":"2026-07-05T09:27:15.980146+00:00"}