{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3B3S7PHHW7A2CLL37GMANGZWUJ","short_pith_number":"pith:3B3S7PHH","schema_version":"1.0","canonical_sha256":"d8772fbce7b7c1a12d7bf998069b36a25480cfb4e8d639f5217b3c588f43a435","source":{"kind":"arxiv","id":"2506.00281","version":1},"attestation_state":"computed","paper":{"title":"Adversarial Threat Vectors and Risk Mitigation for Retrieval-Augmented Generation Systems","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Chris M. Ward, Josh Harguess","submitted_at":"2025-05-30T22:22:05Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) systems, which integrate Large Language Models (LLMs) with external knowledge sources, are vulnerable to a range of adversarial attack vectors. This paper examines the importance of RAG systems through recent industry adoption trends and identifies the prominent attack vectors for RAG: prompt injection, data poisoning, and adversarial query manipulation. We analyze these threats under risk management lens, and propose robust prioritized control list that includes risk-mitigating actions like input validation, adversarial training, and real-time monitoring."},"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":"2506.00281","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2025-05-30T22:22:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3bd745235d2be418da39b178af39305e51dd07c6a9ddac52aa46e85f0ee240be","abstract_canon_sha256":"d65f657f9177c1000311ee1690a824b45ce8f71e200a78b4fd45da728702691f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:32.073059Z","signature_b64":"ugjKuH/Kd7Fl4VjWCUysSpaPvbBeZVPKZb/HBmu34ZI89E8MSWZti6RWqMwqG27BIKeZU9FcqWUSPPbN8/82Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d8772fbce7b7c1a12d7bf998069b36a25480cfb4e8d639f5217b3c588f43a435","last_reissued_at":"2026-07-05T11:13:32.072607Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:32.072607Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adversarial Threat Vectors and Risk Mitigation for Retrieval-Augmented Generation Systems","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Chris M. Ward, Josh Harguess","submitted_at":"2025-05-30T22:22:05Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) systems, which integrate Large Language Models (LLMs) with external knowledge sources, are vulnerable to a range of adversarial attack vectors. This paper examines the importance of RAG systems through recent industry adoption trends and identifies the prominent attack vectors for RAG: prompt injection, data poisoning, and adversarial query manipulation. We analyze these threats under risk management lens, and propose robust prioritized control list that includes risk-mitigating actions like input validation, adversarial training, and real-time monitoring."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.00281","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/2506.00281/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":"2506.00281","created_at":"2026-07-05T11:13:32.072666+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.00281v1","created_at":"2026-07-05T11:13:32.072666+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.00281","created_at":"2026-07-05T11:13:32.072666+00:00"},{"alias_kind":"pith_short_12","alias_value":"3B3S7PHHW7A2","created_at":"2026-07-05T11:13:32.072666+00:00"},{"alias_kind":"pith_short_16","alias_value":"3B3S7PHHW7A2CLL3","created_at":"2026-07-05T11:13:32.072666+00:00"},{"alias_kind":"pith_short_8","alias_value":"3B3S7PHH","created_at":"2026-07-05T11:13:32.072666+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.16014","citing_title":"Ranking Free RAG: Replacing Re-ranking with Selection in RAG for Sensitive Domains","ref_index":9,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ","json":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ.json","graph_json":"https://pith.science/api/pith-number/3B3S7PHHW7A2CLL37GMANGZWUJ/graph.json","events_json":"https://pith.science/api/pith-number/3B3S7PHHW7A2CLL37GMANGZWUJ/events.json","paper":"https://pith.science/paper/3B3S7PHH"},"agent_actions":{"view_html":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ","download_json":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ.json","view_paper":"https://pith.science/paper/3B3S7PHH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.00281&json=true","fetch_graph":"https://pith.science/api/pith-number/3B3S7PHHW7A2CLL37GMANGZWUJ/graph.json","fetch_events":"https://pith.science/api/pith-number/3B3S7PHHW7A2CLL37GMANGZWUJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ/action/storage_attestation","attest_author":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ/action/author_attestation","sign_citation":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ/action/citation_signature","submit_replication":"https://pith.science/pith/3B3S7PHHW7A2CLL37GMANGZWUJ/action/replication_record"}},"created_at":"2026-07-05T11:13:32.072666+00:00","updated_at":"2026-07-05T11:13:32.072666+00:00"}