{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:UVBGDRBQSTVM5Z6O5DFT5CCLB3","short_pith_number":"pith:UVBGDRBQ","schema_version":"1.0","canonical_sha256":"a54261c43094eacee7cee8cb3e884b0ede0fcf2dc5ad4aa4b068a9bee9e21a73","source":{"kind":"arxiv","id":"2607.21094","version":1},"attestation_state":"computed","paper":{"title":"A Polynomial Architecture-Attribution Co-Design Framework for Exact Aumann-Shapley Attribution in GNNs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bizu Feng, Shaode Yu, Shuming Wang, Xiaojun Qian, Yuan Cheng, Zhimu Yang, Zixin Hu","submitted_at":"2026-07-23T09:26:15Z","abstract_excerpt":"We study feature-level and node-level explanations for graph neural networks (GNNs) through the lens of Aumann-Shapley attribution. Path-integral methods such as Integrated Gradients provide an axiomatic formulation of attribution, but their practical use in deep GNNs typically relies on finite-sample numerical approximations to the path integral, requiring a trade-off between quadrature error and computational cost. This paper proposes APEX, a model-attribution co-design framework that makes the attribution integral exactly computable under a polynomial GNN architecture. The key component is "},"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":"2607.21094","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T09:26:15Z","cross_cats_sorted":[],"title_canon_sha256":"2d119d0b6b53c1ff7eb90d02ed1435ccf4d9522da38bc0952075ee9ecfbecc95","abstract_canon_sha256":"63ff4b265639b275a33ae39993086ac64c2ef61c5477370e656ac3ea0867cdfe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T01:23:51.657585Z","signature_b64":"bio4ZT6cBCi0hGDd1VbYVCkZ+KaNpa09y4Z59Gpq94Ap+yOki2S3GgGnHtSch0E3TZNljOE6QHZ0Yr3FJBl9AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a54261c43094eacee7cee8cb3e884b0ede0fcf2dc5ad4aa4b068a9bee9e21a73","last_reissued_at":"2026-07-24T01:23:51.656606Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T01:23:51.656606Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Polynomial Architecture-Attribution Co-Design Framework for Exact Aumann-Shapley Attribution in GNNs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bizu Feng, Shaode Yu, Shuming Wang, Xiaojun Qian, Yuan Cheng, Zhimu Yang, Zixin Hu","submitted_at":"2026-07-23T09:26:15Z","abstract_excerpt":"We study feature-level and node-level explanations for graph neural networks (GNNs) through the lens of Aumann-Shapley attribution. Path-integral methods such as Integrated Gradients provide an axiomatic formulation of attribution, but their practical use in deep GNNs typically relies on finite-sample numerical approximations to the path integral, requiring a trade-off between quadrature error and computational cost. This paper proposes APEX, a model-attribution co-design framework that makes the attribution integral exactly computable under a polynomial GNN architecture. The key component is "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21094","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/2607.21094/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":"2607.21094","created_at":"2026-07-24T01:23:51.657084+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.21094v1","created_at":"2026-07-24T01:23:51.657084+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21094","created_at":"2026-07-24T01:23:51.657084+00:00"},{"alias_kind":"pith_short_12","alias_value":"UVBGDRBQSTVM","created_at":"2026-07-24T01:23:51.657084+00:00"},{"alias_kind":"pith_short_16","alias_value":"UVBGDRBQSTVM5Z6O","created_at":"2026-07-24T01:23:51.657084+00:00"},{"alias_kind":"pith_short_8","alias_value":"UVBGDRBQ","created_at":"2026-07-24T01:23:51.657084+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/UVBGDRBQSTVM5Z6O5DFT5CCLB3","json":"https://pith.science/pith/UVBGDRBQSTVM5Z6O5DFT5CCLB3.json","graph_json":"https://pith.science/api/pith-number/UVBGDRBQSTVM5Z6O5DFT5CCLB3/graph.json","events_json":"https://pith.science/api/pith-number/UVBGDRBQSTVM5Z6O5DFT5CCLB3/events.json","paper":"https://pith.science/paper/UVBGDRBQ"},"agent_actions":{"view_html":"https://pith.science/pith/UVBGDRBQSTVM5Z6O5DFT5CCLB3","download_json":"https://pith.science/pith/UVBGDRBQSTVM5Z6O5DFT5CCLB3.json","view_paper":"https://pith.science/paper/UVBGDRBQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.21094&json=true","fetch_graph":"https://pith.science/api/pith-number/UVBGDRBQSTVM5Z6O5DFT5CCLB3/graph.json","fetch_events":"https://pith.science/api/pith-number/UVBGDRBQSTVM5Z6O5DFT5CCLB3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UVBGDRBQSTVM5Z6O5DFT5CCLB3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UVBGDRBQSTVM5Z6O5DFT5CCLB3/action/storage_attestation","attest_author":"https://pith.science/pith/UVBGDRBQSTVM5Z6O5DFT5CCLB3/action/author_attestation","sign_citation":"https://pith.science/pith/UVBGDRBQSTVM5Z6O5DFT5CCLB3/action/citation_signature","submit_replication":"https://pith.science/pith/UVBGDRBQSTVM5Z6O5DFT5CCLB3/action/replication_record"}},"created_at":"2026-07-24T01:23:51.657084+00:00","updated_at":"2026-07-24T01:23:51.657084+00:00"}