{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OEWIM2OPDD4XUT7WNJT2WZ3WEF","short_pith_number":"pith:OEWIM2OP","schema_version":"1.0","canonical_sha256":"712c8669cf18f97a4ff66a67ab6776215844ca9b9462265b424d1c8024050109","source":{"kind":"arxiv","id":"2502.10673","version":1},"attestation_state":"computed","paper":{"title":"Dataset Protection via Watermarked Canaries in Retrieval-Augmented LLMs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CR","authors_text":"Dawn Song, Xuandong Zhao, Yepeng Liu, Yuheng Bu","submitted_at":"2025-02-15T04:56:45Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) has become an effective method for enhancing large language models (LLMs) with up-to-date knowledge. However, it poses a significant risk of IP infringement, as IP datasets may be incorporated into the knowledge database by malicious Retrieval-Augmented LLMs (RA-LLMs) without authorization. To protect the rights of the dataset owner, an effective dataset membership inference algorithm for RA-LLMs is needed. In this work, we introduce a novel approach to safeguard the ownership of text datasets and effectively detect unauthorized use by the RA-LLMs. Our appr"},"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":"2502.10673","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-15T04:56:45Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5ebae0a9a63ccad872673983b4352b99b54b8d06b60f72a5ac498276ad65e298","abstract_canon_sha256":"b6478fefe321cca3b3f9893cfdc700b3a62cb982bd0a0b22f9ef7c1a0216e17c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:14:56.157630Z","signature_b64":"SEW/vue3pR6Aaee0hjUr2p9iujzmcQfSt4YBksUOzISywUFi3orp6DIxVvdW0JhoBhdX5B5aH/ZInbQ0Y4JSBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"712c8669cf18f97a4ff66a67ab6776215844ca9b9462265b424d1c8024050109","last_reissued_at":"2026-07-05T10:14:56.157144Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:14:56.157144Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dataset Protection via Watermarked Canaries in Retrieval-Augmented LLMs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CR","authors_text":"Dawn Song, Xuandong Zhao, Yepeng Liu, Yuheng Bu","submitted_at":"2025-02-15T04:56:45Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) has become an effective method for enhancing large language models (LLMs) with up-to-date knowledge. However, it poses a significant risk of IP infringement, as IP datasets may be incorporated into the knowledge database by malicious Retrieval-Augmented LLMs (RA-LLMs) without authorization. To protect the rights of the dataset owner, an effective dataset membership inference algorithm for RA-LLMs is needed. In this work, we introduce a novel approach to safeguard the ownership of text datasets and effectively detect unauthorized use by the RA-LLMs. Our appr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.10673","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/2502.10673/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":"2502.10673","created_at":"2026-07-05T10:14:56.157202+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.10673v1","created_at":"2026-07-05T10:14:56.157202+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.10673","created_at":"2026-07-05T10:14:56.157202+00:00"},{"alias_kind":"pith_short_12","alias_value":"OEWIM2OPDD4X","created_at":"2026-07-05T10:14:56.157202+00:00"},{"alias_kind":"pith_short_16","alias_value":"OEWIM2OPDD4XUT7W","created_at":"2026-07-05T10:14:56.157202+00:00"},{"alias_kind":"pith_short_8","alias_value":"OEWIM2OP","created_at":"2026-07-05T10:14:56.157202+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25002","citing_title":"MemMark: State-Evolution Attribution Watermarking for Agent Long-Term Memory Systems","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2510.18333","citing_title":"Position: LLM Watermarking Should Align Stakeholders' Incentives for Practical Adoption","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08826","citing_title":"Fundamental Trade-Offs in Multi-Bit Watermarking of Stochastic Processes","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF","json":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF.json","graph_json":"https://pith.science/api/pith-number/OEWIM2OPDD4XUT7WNJT2WZ3WEF/graph.json","events_json":"https://pith.science/api/pith-number/OEWIM2OPDD4XUT7WNJT2WZ3WEF/events.json","paper":"https://pith.science/paper/OEWIM2OP"},"agent_actions":{"view_html":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF","download_json":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF.json","view_paper":"https://pith.science/paper/OEWIM2OP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.10673&json=true","fetch_graph":"https://pith.science/api/pith-number/OEWIM2OPDD4XUT7WNJT2WZ3WEF/graph.json","fetch_events":"https://pith.science/api/pith-number/OEWIM2OPDD4XUT7WNJT2WZ3WEF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF/action/storage_attestation","attest_author":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF/action/author_attestation","sign_citation":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF/action/citation_signature","submit_replication":"https://pith.science/pith/OEWIM2OPDD4XUT7WNJT2WZ3WEF/action/replication_record"}},"created_at":"2026-07-05T10:14:56.157202+00:00","updated_at":"2026-07-05T10:14:56.157202+00:00"}