{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PDKTABKF2RYNYTRYXJJRDBLITD","short_pith_number":"pith:PDKTABKF","schema_version":"1.0","canonical_sha256":"78d5300545d470dc4e38ba5311856898f51b5ca2e53f3974bd1e7f10a86a29f4","source":{"kind":"arxiv","id":"2506.19109","version":1},"attestation_state":"computed","paper":{"title":"Enhancing Security in LLM Applications: A Performance Evaluation of Early Detection Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Hayretdin Bahsi, Valerii Gakh","submitted_at":"2025-06-23T20:39:43Z","abstract_excerpt":"Prompt injection threatens novel applications that emerge from adapting LLMs for various user tasks. The newly developed LLM-based software applications become more ubiquitous and diverse. However, the threat of prompt injection attacks undermines the security of these systems as the mitigation and defenses against them, proposed so far, are insufficient. We investigated the capabilities of early prompt injection detection systems, focusing specifically on the detection performance of techniques implemented in various open-source solutions. These solutions are supposed to detect certain types "},"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":"2506.19109","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-23T20:39:43Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b1770bfb1028d6c87c65633e229f10c954368ba9b9701a9e5a88196b07cabab7","abstract_canon_sha256":"ef1f2365e694fd944617030476594eec1b3b5ff2a4ee68951b3a30351b4eedcf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:09.594248Z","signature_b64":"H+C/T0MmaqEuguaX5bgytkW+acSmztvHI7tXNh83qUCTp5b8eUSj2b7cwQ83gu6J9YNw/k0dIaw4E0TCEvw7CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"78d5300545d470dc4e38ba5311856898f51b5ca2e53f3974bd1e7f10a86a29f4","last_reissued_at":"2026-07-05T11:26:09.593788Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:09.593788Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Security in LLM Applications: A Performance Evaluation of Early Detection Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Hayretdin Bahsi, Valerii Gakh","submitted_at":"2025-06-23T20:39:43Z","abstract_excerpt":"Prompt injection threatens novel applications that emerge from adapting LLMs for various user tasks. The newly developed LLM-based software applications become more ubiquitous and diverse. However, the threat of prompt injection attacks undermines the security of these systems as the mitigation and defenses against them, proposed so far, are insufficient. We investigated the capabilities of early prompt injection detection systems, focusing specifically on the detection performance of techniques implemented in various open-source solutions. These solutions are supposed to detect certain types "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.19109","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/2506.19109/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":"2506.19109","created_at":"2026-07-05T11:26:09.593847+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.19109v1","created_at":"2026-07-05T11:26:09.593847+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.19109","created_at":"2026-07-05T11:26:09.593847+00:00"},{"alias_kind":"pith_short_12","alias_value":"PDKTABKF2RYN","created_at":"2026-07-05T11:26:09.593847+00:00"},{"alias_kind":"pith_short_16","alias_value":"PDKTABKF2RYNYTRY","created_at":"2026-07-05T11:26:09.593847+00:00"},{"alias_kind":"pith_short_8","alias_value":"PDKTABKF","created_at":"2026-07-05T11:26:09.593847+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07539","citing_title":"Prompt Governance? On Governing Technologies Governed by Natural Language","ref_index":105,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD","json":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD.json","graph_json":"https://pith.science/api/pith-number/PDKTABKF2RYNYTRYXJJRDBLITD/graph.json","events_json":"https://pith.science/api/pith-number/PDKTABKF2RYNYTRYXJJRDBLITD/events.json","paper":"https://pith.science/paper/PDKTABKF"},"agent_actions":{"view_html":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD","download_json":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD.json","view_paper":"https://pith.science/paper/PDKTABKF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.19109&json=true","fetch_graph":"https://pith.science/api/pith-number/PDKTABKF2RYNYTRYXJJRDBLITD/graph.json","fetch_events":"https://pith.science/api/pith-number/PDKTABKF2RYNYTRYXJJRDBLITD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD/action/storage_attestation","attest_author":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD/action/author_attestation","sign_citation":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD/action/citation_signature","submit_replication":"https://pith.science/pith/PDKTABKF2RYNYTRYXJJRDBLITD/action/replication_record"}},"created_at":"2026-07-05T11:26:09.593847+00:00","updated_at":"2026-07-05T11:26:09.593847+00:00"}