{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VLB4AVWQ43SSWXCNR3AEWKSAFX","short_pith_number":"pith:VLB4AVWQ","schema_version":"1.0","canonical_sha256":"aac3c056d0e6e52b5c4d8ec04b2a402de7812bb20636e6646d900321f3e0e57b","source":{"kind":"arxiv","id":"2605.14137","version":1},"attestation_state":"computed","paper":{"title":"Flow Field Reconstruction with Sensor Placement Policy Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A directional transport-aware GNN paired with constrained PPO jointly optimizes flow reconstruction and sensor placement under realistic conditions.","cross_cats":[],"primary_cat":"cs.CE","authors_text":"Guancheng Wan, Haixin Wang, Ruoyan Li, Wei Wang, Xiao Luo, Yizhou Sun, Zijie Huang, Zixiao Liu","submitted_at":"2026-05-13T21:41:29Z","abstract_excerpt":"Flow-field reconstruction from sparse sensor measurements remains a central challenge in modern fluid dynamics, as the need for high-fidelity data often conflicts with practical limits on sensor deployment. Existing deep learning-based methods have demonstrated promising results, but they typically depend on simplifying assumptions such as two-dimensional domains, predefined governing equations, synthetic datasets derived from idealized flow physics, and unconstrained sensor placement. In this work, we address these limitations by studying flow reconstruction under realistic conditions and int"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.14137","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CE","submitted_at":"2026-05-13T21:41:29Z","cross_cats_sorted":[],"title_canon_sha256":"0bbcad029c40a183201c9213f95e2332a0b3dcbbbd031b61e90e4e30c6239cba","abstract_canon_sha256":"8f47ab965166c265c9c5f677836ba09a5c7de41e871c0206ce78100e22764d73"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:11.729883Z","signature_b64":"2M9o8VD0iWALMAq8+5J/VNakNcGabLsMD+1VkjBSgn68dA2awjTAa5lXWGonlMaAKZYau8HGgEb5s0pd5qiBCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aac3c056d0e6e52b5c4d8ec04b2a402de7812bb20636e6646d900321f3e0e57b","last_reissued_at":"2026-05-17T23:39:11.729372Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:11.729372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Flow Field Reconstruction with Sensor Placement Policy Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A directional transport-aware GNN paired with constrained PPO jointly optimizes flow reconstruction and sensor placement under realistic conditions.","cross_cats":[],"primary_cat":"cs.CE","authors_text":"Guancheng Wan, Haixin Wang, Ruoyan Li, Wei Wang, Xiao Luo, Yizhou Sun, Zijie Huang, Zixiao Liu","submitted_at":"2026-05-13T21:41:29Z","abstract_excerpt":"Flow-field reconstruction from sparse sensor measurements remains a central challenge in modern fluid dynamics, as the need for high-fidelity data often conflicts with practical limits on sensor deployment. Existing deep learning-based methods have demonstrated promising results, but they typically depend on simplifying assumptions such as two-dimensional domains, predefined governing equations, synthetic datasets derived from idealized flow physics, and unconstrained sensor placement. In this work, we address these limitations by studying flow reconstruction under realistic conditions and int"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Together, they achieve significant improvements over existing methods.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The directional transport-aware GNN and Two-Step Constrained PPO will maintain performance gains when applied to truly realistic 3D flows without the simplifying assumptions of prior work.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A directional GNN combined with constrained PPO jointly improves flow-field reconstruction accuracy and sensor layout selection in realistic fluid dynamics settings.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A directional transport-aware GNN paired with constrained PPO jointly optimizes flow reconstruction and sensor placement under realistic conditions.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"05c0446c498219a5080cea6ed2732b800865131a48631eecab00a82c529d4fb2"},"source":{"id":"2605.14137","kind":"arxiv","version":1},"verdict":{"id":"c631fc91-b64d-4c39-be1a-9dd56dde257e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T01:55:18.412476Z","strongest_claim":"Together, they achieve significant improvements over existing methods.","one_line_summary":"A directional GNN combined with constrained PPO jointly improves flow-field reconstruction accuracy and sensor layout selection in realistic fluid dynamics settings.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The directional transport-aware GNN and Two-Step Constrained PPO will maintain performance gains when applied to truly realistic 3D flows without the simplifying assumptions of prior work.","pith_extraction_headline":"A directional transport-aware GNN paired with constrained PPO jointly optimizes flow reconstruction and sensor placement under realistic conditions."},"references":{"count":26,"sample":[{"doi":"10.1002/aic.690480510","year":null,"title":"COMSOL Multiphysics®","work_id":"93e41390-3f44-4d26-87a9-6555afc28687","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Asymptotic accuracy of the saddlepoint approximation for maximum likelihood estimation.The Annals of Statistics, 50(4):2021–2046,","work_id":"78b2a777-8d13-4522-bbdd-0e027067061d","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1214/22-aos2169","year":null,"title":"URL https://doi.org/10.1214/22-AOS2169","work_id":"a8f24731-e95a-4b61-b774-b2857f5b9933","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1063/1.2393436","year":null,"title":"doi: 10.1063/1.2393436","work_id":"96758ac0-736d-40a3-931f-fa8f21b91195","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1063/5.0097688","year":null,"title":"doi: 10.1063/5.0097688","work_id":"dcc92603-0835-4985-8267-c34090c31c70","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":26,"snapshot_sha256":"98f453b555ea9e08b773df72c624ae237fd38b1fb6d083469f6da6119f1b2782","internal_anchors":2},"formal_canon":{"evidence_count":2,"snapshot_sha256":"060be1789accc820da85ee3fb41ad8ff2e40fe3e6fb6b757aba25248074b1e1b"},"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":"2605.14137","created_at":"2026-05-17T23:39:11.729454+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.14137v1","created_at":"2026-05-17T23:39:11.729454+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14137","created_at":"2026-05-17T23:39:11.729454+00:00"},{"alias_kind":"pith_short_12","alias_value":"VLB4AVWQ43SS","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"VLB4AVWQ43SSWXCN","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"VLB4AVWQ","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX","json":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX.json","graph_json":"https://pith.science/api/pith-number/VLB4AVWQ43SSWXCNR3AEWKSAFX/graph.json","events_json":"https://pith.science/api/pith-number/VLB4AVWQ43SSWXCNR3AEWKSAFX/events.json","paper":"https://pith.science/paper/VLB4AVWQ"},"agent_actions":{"view_html":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX","download_json":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX.json","view_paper":"https://pith.science/paper/VLB4AVWQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.14137&json=true","fetch_graph":"https://pith.science/api/pith-number/VLB4AVWQ43SSWXCNR3AEWKSAFX/graph.json","fetch_events":"https://pith.science/api/pith-number/VLB4AVWQ43SSWXCNR3AEWKSAFX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX/action/storage_attestation","attest_author":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX/action/author_attestation","sign_citation":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX/action/citation_signature","submit_replication":"https://pith.science/pith/VLB4AVWQ43SSWXCNR3AEWKSAFX/action/replication_record"}},"created_at":"2026-05-17T23:39:11.729454+00:00","updated_at":"2026-05-17T23:39:11.729454+00:00"}