{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:K4NBAZ4MXJORCAVE7PXC3CN4VY","short_pith_number":"pith:K4NBAZ4M","schema_version":"1.0","canonical_sha256":"571a10678cba5d1102a4fbee2d89bcae022adf1b0fac40b00f5927db9bf993e3","source":{"kind":"arxiv","id":"2402.04435","version":2},"attestation_state":"computed","paper":{"title":"PreGIP: Watermarking the Pretraining of Graph Neural Networks for Deep Intellectual Property Protection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Enyan Dai, Minhua Lin, Suhang Wang","submitted_at":"2024-02-06T22:13:49Z","abstract_excerpt":"Pretraining on Graph Neural Networks (GNNs) has shown great power in facilitating various downstream tasks. As pretraining generally requires huge amount of data and computational resources, the pretrained GNNs are high-value Intellectual Properties (IP) of the legitimate owner. However, adversaries may illegally copy and deploy the pretrained GNN models for their downstream tasks. Though initial efforts have been made to watermark GNN classifiers for IP protection, these methods require the target classification task for watermarking, and thus are not applicable to self-supervised pretraining"},"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":"2402.04435","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-06T22:13:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d66e940e42a6d8cbe8f3e9c021356bbfb3b72ea622f096372b1858bedbd3ff87","abstract_canon_sha256":"82f263fbaf445364bcc78d00e9fc1ac2686d7a3f298720420fb880e7ea6122a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:07.282101Z","signature_b64":"et8ny3/S2DccTGX/WliBqXH8JK++KbVUNsM94gp/NB/vxFibVlZBesZp3o8shTGDU9qEkL7q5jDQLZJ6W/dmAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"571a10678cba5d1102a4fbee2d89bcae022adf1b0fac40b00f5927db9bf993e3","last_reissued_at":"2026-07-05T11:13:07.281577Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:07.281577Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PreGIP: Watermarking the Pretraining of Graph Neural Networks for Deep Intellectual Property Protection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Enyan Dai, Minhua Lin, Suhang Wang","submitted_at":"2024-02-06T22:13:49Z","abstract_excerpt":"Pretraining on Graph Neural Networks (GNNs) has shown great power in facilitating various downstream tasks. As pretraining generally requires huge amount of data and computational resources, the pretrained GNNs are high-value Intellectual Properties (IP) of the legitimate owner. However, adversaries may illegally copy and deploy the pretrained GNN models for their downstream tasks. Though initial efforts have been made to watermark GNN classifiers for IP protection, these methods require the target classification task for watermarking, and thus are not applicable to self-supervised pretraining"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.04435","kind":"arxiv","version":2},"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/2402.04435/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":"2402.04435","created_at":"2026-07-05T11:13:07.281644+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.04435v2","created_at":"2026-07-05T11:13:07.281644+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.04435","created_at":"2026-07-05T11:13:07.281644+00:00"},{"alias_kind":"pith_short_12","alias_value":"K4NBAZ4MXJOR","created_at":"2026-07-05T11:13:07.281644+00:00"},{"alias_kind":"pith_short_16","alias_value":"K4NBAZ4MXJORCAVE","created_at":"2026-07-05T11:13:07.281644+00:00"},{"alias_kind":"pith_short_8","alias_value":"K4NBAZ4M","created_at":"2026-07-05T11:13:07.281644+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.12827","citing_title":"GraphIP-Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12827","citing_title":"GraphIP-Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY","json":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY.json","graph_json":"https://pith.science/api/pith-number/K4NBAZ4MXJORCAVE7PXC3CN4VY/graph.json","events_json":"https://pith.science/api/pith-number/K4NBAZ4MXJORCAVE7PXC3CN4VY/events.json","paper":"https://pith.science/paper/K4NBAZ4M"},"agent_actions":{"view_html":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY","download_json":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY.json","view_paper":"https://pith.science/paper/K4NBAZ4M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.04435&json=true","fetch_graph":"https://pith.science/api/pith-number/K4NBAZ4MXJORCAVE7PXC3CN4VY/graph.json","fetch_events":"https://pith.science/api/pith-number/K4NBAZ4MXJORCAVE7PXC3CN4VY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY/action/storage_attestation","attest_author":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY/action/author_attestation","sign_citation":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY/action/citation_signature","submit_replication":"https://pith.science/pith/K4NBAZ4MXJORCAVE7PXC3CN4VY/action/replication_record"}},"created_at":"2026-07-05T11:13:07.281644+00:00","updated_at":"2026-07-05T11:13:07.281644+00:00"}