{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LIVR2BBCNLP76ABFYTUNEORUDP","short_pith_number":"pith:LIVR2BBC","schema_version":"1.0","canonical_sha256":"5a2b1d04226adfff0025c4e8d23a341bfdce2b27cf38a9ec90b5568d99422eea","source":{"kind":"arxiv","id":"2504.02458","version":1},"attestation_state":"computed","paper":{"title":"Retrieval-Augmented Purifier for Robust LLM-Empowered Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Liangbo Ning, Qing Li, Wenqi Fan","submitted_at":"2025-04-03T10:22:30Z","abstract_excerpt":"Recently, Large Language Model (LLM)-empowered recommender systems have revolutionized personalized recommendation frameworks and attracted extensive attention. Despite the remarkable success, existing LLM-empowered RecSys have been demonstrated to be highly vulnerable to minor perturbations. To mitigate the negative impact of such vulnerabilities, one potential solution is to employ collaborative signals based on item-item co-occurrence to purify the malicious collaborative knowledge from the user's historical interactions inserted by attackers. On the other hand, due to the capabilities to e"},"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":"2504.02458","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2025-04-03T10:22:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a9c021b4284fc74be3e963ff50d86720069262383b790da7ea9b7d6d721b255e","abstract_canon_sha256":"dffdb7dddcf5633b545571139cf794b3f8d78196abe1e7795cfcb5e10325ddfd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:43:57.449241Z","signature_b64":"yZN6vf8Kwth81qFJ5emQlQyGlfHQfTux+iQFqHdRJB8yAKYFzealaqpfyFdV255E4DAajl2dnJztDHTtzsLuCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a2b1d04226adfff0025c4e8d23a341bfdce2b27cf38a9ec90b5568d99422eea","last_reissued_at":"2026-07-05T10:43:57.448714Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:43:57.448714Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Retrieval-Augmented Purifier for Robust LLM-Empowered Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Liangbo Ning, Qing Li, Wenqi Fan","submitted_at":"2025-04-03T10:22:30Z","abstract_excerpt":"Recently, Large Language Model (LLM)-empowered recommender systems have revolutionized personalized recommendation frameworks and attracted extensive attention. Despite the remarkable success, existing LLM-empowered RecSys have been demonstrated to be highly vulnerable to minor perturbations. To mitigate the negative impact of such vulnerabilities, one potential solution is to employ collaborative signals based on item-item co-occurrence to purify the malicious collaborative knowledge from the user's historical interactions inserted by attackers. On the other hand, due to the capabilities to e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.02458","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/2504.02458/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":"2504.02458","created_at":"2026-07-05T10:43:57.448768+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.02458v1","created_at":"2026-07-05T10:43:57.448768+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.02458","created_at":"2026-07-05T10:43:57.448768+00:00"},{"alias_kind":"pith_short_12","alias_value":"LIVR2BBCNLP7","created_at":"2026-07-05T10:43:57.448768+00:00"},{"alias_kind":"pith_short_16","alias_value":"LIVR2BBCNLP76ABF","created_at":"2026-07-05T10:43:57.448768+00:00"},{"alias_kind":"pith_short_8","alias_value":"LIVR2BBC","created_at":"2026-07-05T10:43:57.448768+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00540","citing_title":"Trustworthy Recommendation in the Era of Large Language Models: Opportunities and Challenges","ref_index":174,"is_internal_anchor":false},{"citing_arxiv_id":"2510.27157","citing_title":"A Survey on Generative Recommendation: Data, Model, and Tasks","ref_index":128,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP","json":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP.json","graph_json":"https://pith.science/api/pith-number/LIVR2BBCNLP76ABFYTUNEORUDP/graph.json","events_json":"https://pith.science/api/pith-number/LIVR2BBCNLP76ABFYTUNEORUDP/events.json","paper":"https://pith.science/paper/LIVR2BBC"},"agent_actions":{"view_html":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP","download_json":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP.json","view_paper":"https://pith.science/paper/LIVR2BBC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.02458&json=true","fetch_graph":"https://pith.science/api/pith-number/LIVR2BBCNLP76ABFYTUNEORUDP/graph.json","fetch_events":"https://pith.science/api/pith-number/LIVR2BBCNLP76ABFYTUNEORUDP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP/action/storage_attestation","attest_author":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP/action/author_attestation","sign_citation":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP/action/citation_signature","submit_replication":"https://pith.science/pith/LIVR2BBCNLP76ABFYTUNEORUDP/action/replication_record"}},"created_at":"2026-07-05T10:43:57.448768+00:00","updated_at":"2026-07-05T10:43:57.448768+00:00"}