{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YAQ7KWWRSNXPUYU5CDNYH6X6TB","short_pith_number":"pith:YAQ7KWWR","schema_version":"1.0","canonical_sha256":"c021f55ad1936efa629d10db83fafe9854ad6197d276564269060c93d124bec8","source":{"kind":"arxiv","id":"2506.12558","version":1},"attestation_state":"computed","paper":{"title":"RAW-Explainer: Post-hoc Explanations of Graph Neural Networks on Knowledge Graphs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Djellel Difallah, Ryoji Kubo","submitted_at":"2025-06-14T15:55:17Z","abstract_excerpt":"Graph neural networks have demonstrated state-of-the-art performance on knowledge graph tasks such as link prediction. However, interpreting GNN predictions remains a challenging open problem. While many GNN explainability methods have been proposed for node or graph-level tasks, approaches for generating explanations for link predictions in heterogeneous settings are limited. In this paper, we propose RAW-Explainer, a novel framework designed to generate connected, concise, and thus interpretable subgraph explanations for link prediction. Our method leverages the heterogeneous information in "},"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":"2506.12558","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-14T15:55:17Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"eddb802987f3c2e150ae10993663751b3daa44f57353d5dd7ffc03b7cb68d6b6","abstract_canon_sha256":"2cdbc88777367d08cf718f8023449c28e1a883898a75b6346094535ff845935e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:55.278856Z","signature_b64":"ztiO5BxzihQGjTVMinvGGHO1IiBbRovhhKmVdY8UwUWc2fRlxXedl1JhB9pUgifP+07jZkPd7o4W9QaGR1EbAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c021f55ad1936efa629d10db83fafe9854ad6197d276564269060c93d124bec8","last_reissued_at":"2026-07-05T11:21:55.278343Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:55.278343Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RAW-Explainer: Post-hoc Explanations of Graph Neural Networks on Knowledge Graphs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Djellel Difallah, Ryoji Kubo","submitted_at":"2025-06-14T15:55:17Z","abstract_excerpt":"Graph neural networks have demonstrated state-of-the-art performance on knowledge graph tasks such as link prediction. However, interpreting GNN predictions remains a challenging open problem. While many GNN explainability methods have been proposed for node or graph-level tasks, approaches for generating explanations for link predictions in heterogeneous settings are limited. In this paper, we propose RAW-Explainer, a novel framework designed to generate connected, concise, and thus interpretable subgraph explanations for link prediction. Our method leverages the heterogeneous information in "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.12558","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/2506.12558/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":"2506.12558","created_at":"2026-07-05T11:21:55.278405+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.12558v1","created_at":"2026-07-05T11:21:55.278405+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.12558","created_at":"2026-07-05T11:21:55.278405+00:00"},{"alias_kind":"pith_short_12","alias_value":"YAQ7KWWRSNXP","created_at":"2026-07-05T11:21:55.278405+00:00"},{"alias_kind":"pith_short_16","alias_value":"YAQ7KWWRSNXPUYU5","created_at":"2026-07-05T11:21:55.278405+00:00"},{"alias_kind":"pith_short_8","alias_value":"YAQ7KWWR","created_at":"2026-07-05T11:21:55.278405+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB","json":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB.json","graph_json":"https://pith.science/api/pith-number/YAQ7KWWRSNXPUYU5CDNYH6X6TB/graph.json","events_json":"https://pith.science/api/pith-number/YAQ7KWWRSNXPUYU5CDNYH6X6TB/events.json","paper":"https://pith.science/paper/YAQ7KWWR"},"agent_actions":{"view_html":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB","download_json":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB.json","view_paper":"https://pith.science/paper/YAQ7KWWR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.12558&json=true","fetch_graph":"https://pith.science/api/pith-number/YAQ7KWWRSNXPUYU5CDNYH6X6TB/graph.json","fetch_events":"https://pith.science/api/pith-number/YAQ7KWWRSNXPUYU5CDNYH6X6TB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB/action/storage_attestation","attest_author":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB/action/author_attestation","sign_citation":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB/action/citation_signature","submit_replication":"https://pith.science/pith/YAQ7KWWRSNXPUYU5CDNYH6X6TB/action/replication_record"}},"created_at":"2026-07-05T11:21:55.278405+00:00","updated_at":"2026-07-05T11:21:55.278405+00:00"}