{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4CZVSHLDBSKWZTIDQP3BIWSXAQ","short_pith_number":"pith:4CZVSHLD","schema_version":"1.0","canonical_sha256":"e0b3591d630c956ccd0383f6145a570430a62cbf5c741ba9f7f4fff1f4bc2b45","source":{"kind":"arxiv","id":"2506.11398","version":1},"attestation_state":"computed","paper":{"title":"FIGNN: Feature-Specific Interpretability for Graph Neural Network Surrogate Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["physics.flu-dyn"],"primary_cat":"cs.LG","authors_text":"Riddhiman Raut, Romit Maulik, Shivam Barwey","submitted_at":"2025-06-13T01:45:37Z","abstract_excerpt":"This work presents a novel graph neural network (GNN) architecture, the Feature-specific Interpretable Graph Neural Network (FIGNN), designed to enhance the interpretability of deep learning surrogate models defined on unstructured grids in scientific applications. Traditional GNNs often obscure the distinct spatial influences of different features in multivariate prediction tasks. FIGNN addresses this limitation by introducing a feature-specific pooling strategy, which enables independent attribution of spatial importance for each predicted variable. Additionally, a mask-based regularization "},"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":"2506.11398","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-13T01:45:37Z","cross_cats_sorted":["physics.flu-dyn"],"title_canon_sha256":"41db1ca1f046d071d7966e2313830860320b1860c8dc808d5532eea65109ac49","abstract_canon_sha256":"6791ac18ee75ec44dc2728709254139fcfa4f72782855a3d3b65e66f996a7729"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:03.641602Z","signature_b64":"GZdggTDAtSnk4OVqvi045avAFujiWhTXXfWXNLsQMbTYsdXo0Mj/Vnd9rYlQqmeY5zYOyCv8nMsTJ/csTY37BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e0b3591d630c956ccd0383f6145a570430a62cbf5c741ba9f7f4fff1f4bc2b45","last_reissued_at":"2026-07-05T11:21:03.641084Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:03.641084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FIGNN: Feature-Specific Interpretability for Graph Neural Network Surrogate Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["physics.flu-dyn"],"primary_cat":"cs.LG","authors_text":"Riddhiman Raut, Romit Maulik, Shivam Barwey","submitted_at":"2025-06-13T01:45:37Z","abstract_excerpt":"This work presents a novel graph neural network (GNN) architecture, the Feature-specific Interpretable Graph Neural Network (FIGNN), designed to enhance the interpretability of deep learning surrogate models defined on unstructured grids in scientific applications. Traditional GNNs often obscure the distinct spatial influences of different features in multivariate prediction tasks. FIGNN addresses this limitation by introducing a feature-specific pooling strategy, which enables independent attribution of spatial importance for each predicted variable. Additionally, a mask-based regularization "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11398","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/2506.11398/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":"2506.11398","created_at":"2026-07-05T11:21:03.641139+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11398v1","created_at":"2026-07-05T11:21:03.641139+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11398","created_at":"2026-07-05T11:21:03.641139+00:00"},{"alias_kind":"pith_short_12","alias_value":"4CZVSHLDBSKW","created_at":"2026-07-05T11:21:03.641139+00:00"},{"alias_kind":"pith_short_16","alias_value":"4CZVSHLDBSKWZTID","created_at":"2026-07-05T11:21:03.641139+00:00"},{"alias_kind":"pith_short_8","alias_value":"4CZVSHLD","created_at":"2026-07-05T11:21:03.641139+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25115","citing_title":"Courant: a State-Adaptive Perceiver-Based Neural Surrogate with Local Support and Interpretable Field Decomposition","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ","json":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ.json","graph_json":"https://pith.science/api/pith-number/4CZVSHLDBSKWZTIDQP3BIWSXAQ/graph.json","events_json":"https://pith.science/api/pith-number/4CZVSHLDBSKWZTIDQP3BIWSXAQ/events.json","paper":"https://pith.science/paper/4CZVSHLD"},"agent_actions":{"view_html":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ","download_json":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ.json","view_paper":"https://pith.science/paper/4CZVSHLD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11398&json=true","fetch_graph":"https://pith.science/api/pith-number/4CZVSHLDBSKWZTIDQP3BIWSXAQ/graph.json","fetch_events":"https://pith.science/api/pith-number/4CZVSHLDBSKWZTIDQP3BIWSXAQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ/action/storage_attestation","attest_author":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ/action/author_attestation","sign_citation":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ/action/citation_signature","submit_replication":"https://pith.science/pith/4CZVSHLDBSKWZTIDQP3BIWSXAQ/action/replication_record"}},"created_at":"2026-07-05T11:21:03.641139+00:00","updated_at":"2026-07-05T11:21:03.641139+00:00"}