{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KQYOUUYCU65HXNWVEQR7D2L3EG","short_pith_number":"pith:KQYOUUYC","schema_version":"1.0","canonical_sha256":"5430ea5302a7ba7bb6d52423f1e97b2194ca17b00c0caeba353a8514c4d7f207","source":{"kind":"arxiv","id":"2501.02968","version":7},"attestation_state":"computed","paper":{"title":"FlippedRAG: Black-Box Opinion Manipulation Adversarial Attacks to Retrieval-Augmented Generation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Fan Zhang, Haotan Liu, Jiawei Liu, Miaokun Chen, Qikai Cheng, Wei Lu, Xiaozhong Liu, Yuyang Gong, Zhuo Chen","submitted_at":"2025-01-06T12:24:57Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) enriches LLMs by dynamically retrieving external knowledge, reducing hallucinations and satisfying real-time information needs. While existing research mainly targets RAG's performance and efficiency, emerging studies highlight critical security concerns. Yet, current adversarial approaches remain limited, mostly addressing white-box scenarios or heuristic black-box attacks without fully investigating vulnerabilities in the retrieval phase. Additionally, prior works mainly focus on factoid Q&A tasks, their attacks lack complexity and can be easily corrected"},"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":"2501.02968","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2025-01-06T12:24:57Z","cross_cats_sorted":[],"title_canon_sha256":"626588021bfa2888412ef1771b119f940605ffd63edb97abd7e873959cb5506b","abstract_canon_sha256":"72c054b6ce482785e058b9e5d2ed8551c0a0060b284a9b9681d1600ef528f943"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T02:18:11.553320Z","signature_b64":"qn8Vgz8ONBiyPMZeVwSgwC/RWEU5Mx4urKuLz7F5Ulb4Je+1MFR75gAsTM70LiTgmYJ9bffUdG4eqnNwjTR9Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5430ea5302a7ba7bb6d52423f1e97b2194ca17b00c0caeba353a8514c4d7f207","last_reissued_at":"2026-08-11T02:18:11.551186Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T02:18:11.551186Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FlippedRAG: Black-Box Opinion Manipulation Adversarial Attacks to Retrieval-Augmented Generation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Fan Zhang, Haotan Liu, Jiawei Liu, Miaokun Chen, Qikai Cheng, Wei Lu, Xiaozhong Liu, Yuyang Gong, Zhuo Chen","submitted_at":"2025-01-06T12:24:57Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) enriches LLMs by dynamically retrieving external knowledge, reducing hallucinations and satisfying real-time information needs. While existing research mainly targets RAG's performance and efficiency, emerging studies highlight critical security concerns. Yet, current adversarial approaches remain limited, mostly addressing white-box scenarios or heuristic black-box attacks without fully investigating vulnerabilities in the retrieval phase. Additionally, prior works mainly focus on factoid Q&A tasks, their attacks lack complexity and can be easily corrected"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02968","kind":"arxiv","version":7},"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/2501.02968/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":"2501.02968","created_at":"2026-08-11T02:18:11.552021+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.02968v7","created_at":"2026-08-11T02:18:11.552021+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02968","created_at":"2026-08-11T02:18:11.552021+00:00"},{"alias_kind":"pith_short_12","alias_value":"KQYOUUYCU65H","created_at":"2026-08-11T02:18:11.552021+00:00"},{"alias_kind":"pith_short_16","alias_value":"KQYOUUYCU65HXNWV","created_at":"2026-08-11T02:18:11.552021+00:00"},{"alias_kind":"pith_short_8","alias_value":"KQYOUUYC","created_at":"2026-08-11T02:18:11.552021+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.12201","citing_title":"AdversarialCoT: Single-Document Retrieval Poisoning for LLM Reasoning","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG","json":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG.json","graph_json":"https://pith.science/api/pith-number/KQYOUUYCU65HXNWVEQR7D2L3EG/graph.json","events_json":"https://pith.science/api/pith-number/KQYOUUYCU65HXNWVEQR7D2L3EG/events.json","paper":"https://pith.science/paper/KQYOUUYC"},"agent_actions":{"view_html":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG","download_json":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG.json","view_paper":"https://pith.science/paper/KQYOUUYC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.02968&json=true","fetch_graph":"https://pith.science/api/pith-number/KQYOUUYCU65HXNWVEQR7D2L3EG/graph.json","fetch_events":"https://pith.science/api/pith-number/KQYOUUYCU65HXNWVEQR7D2L3EG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG/action/storage_attestation","attest_author":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG/action/author_attestation","sign_citation":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG/action/citation_signature","submit_replication":"https://pith.science/pith/KQYOUUYCU65HXNWVEQR7D2L3EG/action/replication_record"}},"created_at":"2026-08-11T02:18:11.552021+00:00","updated_at":"2026-08-11T02:18:11.552021+00:00"}