{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DVEKH3HK72TPD3TGNAWSK4PTKB","short_pith_number":"pith:DVEKH3HK","schema_version":"1.0","canonical_sha256":"1d48a3eceafea6f1ee66682d2571f3507bbbca90f62b5b76cf49166b94a3d3dc","source":{"kind":"arxiv","id":"2604.08453","version":2},"attestation_state":"computed","paper":{"title":"Hard-constrained Physics-informed Neural Networks for Interface Problems","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Hard-constrained PINN formulations embed interface continuity and flux conditions directly into the neural network solution representation.","cross_cats":["cs.NA","physics.comp-ph"],"primary_cat":"math.NA","authors_text":"Michael S. Penwarden, Pratanu Roy, Seung Whan Chung, Stephen T. Castonguay, Sumanta Roy, Yucheng Fu","submitted_at":"2026-04-09T16:49:55Z","abstract_excerpt":"Physics-informed neural networks (PINNs) have emerged as a flexible framework for solving partial differential equations, but their performance on interface problems remains challenging because continuity and flux conditions are typically imposed through soft penalty terms. The standard soft-constraint formulation leads to imperfect interface enforcement and degraded accuracy near interfaces. We introduce two ansatz-based hard-constrained PINN formulations for interface problems that embed the interface physics into the solution representation and thereby decouple interface enforcement from PD"},"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":"2604.08453","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"math.NA","submitted_at":"2026-04-09T16:49:55Z","cross_cats_sorted":["cs.NA","physics.comp-ph"],"title_canon_sha256":"c6156e3b3572c1373c9702aa0d0968f355301b7bc8194517c2970de6079def09","abstract_canon_sha256":"d2965726b33b9ad8d3a2388fb75403578ce05707475198fc25d0b25308903570"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:11.354438Z","signature_b64":"IccyGz4EoxLOYXlN9EfabBVcqQVj9roM5WLsthjV3qkAAihWFzkTA5laOYceyFppIeg5kWOB2jvp3GYT8n1qBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d48a3eceafea6f1ee66682d2571f3507bbbca90f62b5b76cf49166b94a3d3dc","last_reissued_at":"2026-05-20T00:02:11.353662Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:11.353662Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hard-constrained Physics-informed Neural Networks for Interface Problems","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Hard-constrained PINN formulations embed interface continuity and flux conditions directly into the neural network solution representation.","cross_cats":["cs.NA","physics.comp-ph"],"primary_cat":"math.NA","authors_text":"Michael S. Penwarden, Pratanu Roy, Seung Whan Chung, Stephen T. Castonguay, Sumanta Roy, Yucheng Fu","submitted_at":"2026-04-09T16:49:55Z","abstract_excerpt":"Physics-informed neural networks (PINNs) have emerged as a flexible framework for solving partial differential equations, but their performance on interface problems remains challenging because continuity and flux conditions are typically imposed through soft penalty terms. The standard soft-constraint formulation leads to imperfect interface enforcement and degraded accuracy near interfaces. We introduce two ansatz-based hard-constrained PINN formulations for interface problems that embed the interface physics into the solution representation and thereby decouple interface enforcement from PD"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We introduce two ansatz-based hard-constrained PINN formulations for interface problems that embed the interface physics into the solution representation and thereby decouple interface enforcement from PDE residual minimization.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the windowing construction remains stable for general interface geometries and that the discrete buffer correction points can be chosen without introducing new fitting parameters that affect accuracy.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Hard-constrained PINN formulations via windowing and buffer approaches enforce interface conditions by design and outperform soft-constrained baselines on 1D and 2D elliptic interface problems.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Hard-constrained PINN formulations embed interface continuity and flux conditions directly into the neural network solution representation.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"bf6c4ac925f166c902ea89ad9b28681e7472aa6068d71d8645619cf37d2718a7"},"source":{"id":"2604.08453","kind":"arxiv","version":2},"verdict":{"id":"665b6f84-fd08-4a9e-bf75-25015d7e1ba2","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T17:04:06.400202Z","strongest_claim":"We introduce two ansatz-based hard-constrained PINN formulations for interface problems that embed the interface physics into the solution representation and thereby decouple interface enforcement from PDE residual minimization.","one_line_summary":"Hard-constrained PINN formulations via windowing and buffer approaches enforce interface conditions by design and outperform soft-constrained baselines on 1D and 2D elliptic interface problems.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the windowing construction remains stable for general interface geometries and that the discrete buffer correction points can be chosen without introducing new fitting parameters that affect accuracy.","pith_extraction_headline":"Hard-constrained PINN formulations embed interface continuity and flux conditions directly into the neural network solution representation."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.08453/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":"2604.08453","created_at":"2026-05-20T00:02:11.353772+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.08453v2","created_at":"2026-05-20T00:02:11.353772+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.08453","created_at":"2026-05-20T00:02:11.353772+00:00"},{"alias_kind":"pith_short_12","alias_value":"DVEKH3HK72TP","created_at":"2026-05-20T00:02:11.353772+00:00"},{"alias_kind":"pith_short_16","alias_value":"DVEKH3HK72TPD3TG","created_at":"2026-05-20T00:02:11.353772+00:00"},{"alias_kind":"pith_short_8","alias_value":"DVEKH3HK","created_at":"2026-05-20T00:02:11.353772+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/DVEKH3HK72TPD3TGNAWSK4PTKB","json":"https://pith.science/pith/DVEKH3HK72TPD3TGNAWSK4PTKB.json","graph_json":"https://pith.science/api/pith-number/DVEKH3HK72TPD3TGNAWSK4PTKB/graph.json","events_json":"https://pith.science/api/pith-number/DVEKH3HK72TPD3TGNAWSK4PTKB/events.json","paper":"https://pith.science/paper/DVEKH3HK"},"agent_actions":{"view_html":"https://pith.science/pith/DVEKH3HK72TPD3TGNAWSK4PTKB","download_json":"https://pith.science/pith/DVEKH3HK72TPD3TGNAWSK4PTKB.json","view_paper":"https://pith.science/paper/DVEKH3HK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.08453&json=true","fetch_graph":"https://pith.science/api/pith-number/DVEKH3HK72TPD3TGNAWSK4PTKB/graph.json","fetch_events":"https://pith.science/api/pith-number/DVEKH3HK72TPD3TGNAWSK4PTKB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DVEKH3HK72TPD3TGNAWSK4PTKB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DVEKH3HK72TPD3TGNAWSK4PTKB/action/storage_attestation","attest_author":"https://pith.science/pith/DVEKH3HK72TPD3TGNAWSK4PTKB/action/author_attestation","sign_citation":"https://pith.science/pith/DVEKH3HK72TPD3TGNAWSK4PTKB/action/citation_signature","submit_replication":"https://pith.science/pith/DVEKH3HK72TPD3TGNAWSK4PTKB/action/replication_record"}},"created_at":"2026-05-20T00:02:11.353772+00:00","updated_at":"2026-05-20T00:02:11.353772+00:00"}