{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RFSJ3COXIGHRBEOOZBYVMCGL3J","short_pith_number":"pith:RFSJ3COX","schema_version":"1.0","canonical_sha256":"89649d89d7418f1091cec8715608cbda7b68985b7115f2b506fdf78f08932a77","source":{"kind":"arxiv","id":"2506.05484","version":1},"attestation_state":"computed","paper":{"title":"Initial Model Incorporation for Deep Learning FWI: Pretraining or Denormalization?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["physics.geo-ph"],"primary_cat":"cs.LG","authors_text":"Bangyu Wu, Kai Yang, Meng Li, Ruihua Chen","submitted_at":"2025-06-05T18:06:37Z","abstract_excerpt":"Subsurface property neural network reparameterized full waveform inversion (FWI) has emerged as an effective unsupervised learning framework, which can invert stably with an inaccurate starting model. It updates the trainable neural network parameters instead of fine-tuning on the subsurface model directly. There are primarily two ways to embed the prior knowledge of the initial model into neural networks, that is, pretraining and denormalization. Pretraining first regulates the neural networks' parameters by fitting the initial velocity model; Denormalization directly adds the outputs of the "},"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":"2506.05484","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T18:06:37Z","cross_cats_sorted":["physics.geo-ph"],"title_canon_sha256":"40b1f6ea971bbc261f93b5fd345082ee2709945536a0fe7fdd4b00f1c1e34c20","abstract_canon_sha256":"f6a79ed26f4f16ae9d534dd8e68ea1c69b51eb7debca26efefcb1adbf5367c21"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:17:03.452446Z","signature_b64":"duRY4FkToNlRIy9YBKwH9LcD9IhEAAcpkdcMZcS6v/d9G52r1Agw9jBRtu9Z9a09tqTWkybBCz6fLdH65UNyBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89649d89d7418f1091cec8715608cbda7b68985b7115f2b506fdf78f08932a77","last_reissued_at":"2026-07-05T11:17:03.451916Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:17:03.451916Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Initial Model Incorporation for Deep Learning FWI: Pretraining or Denormalization?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["physics.geo-ph"],"primary_cat":"cs.LG","authors_text":"Bangyu Wu, Kai Yang, Meng Li, Ruihua Chen","submitted_at":"2025-06-05T18:06:37Z","abstract_excerpt":"Subsurface property neural network reparameterized full waveform inversion (FWI) has emerged as an effective unsupervised learning framework, which can invert stably with an inaccurate starting model. It updates the trainable neural network parameters instead of fine-tuning on the subsurface model directly. There are primarily two ways to embed the prior knowledge of the initial model into neural networks, that is, pretraining and denormalization. Pretraining first regulates the neural networks' parameters by fitting the initial velocity model; Denormalization directly adds the outputs of the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05484","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/2506.05484/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":"2506.05484","created_at":"2026-07-05T11:17:03.451977+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.05484v1","created_at":"2026-07-05T11:17:03.451977+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05484","created_at":"2026-07-05T11:17:03.451977+00:00"},{"alias_kind":"pith_short_12","alias_value":"RFSJ3COXIGHR","created_at":"2026-07-05T11:17:03.451977+00:00"},{"alias_kind":"pith_short_16","alias_value":"RFSJ3COXIGHRBEOO","created_at":"2026-07-05T11:17:03.451977+00:00"},{"alias_kind":"pith_short_8","alias_value":"RFSJ3COX","created_at":"2026-07-05T11:17:03.451977+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14370","citing_title":"Deciphering Neural Reparameterized Full-Waveform Inversion with Neural Sensitivity Kernel and Wave Tangent Kernel","ref_index":137,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J","json":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J.json","graph_json":"https://pith.science/api/pith-number/RFSJ3COXIGHRBEOOZBYVMCGL3J/graph.json","events_json":"https://pith.science/api/pith-number/RFSJ3COXIGHRBEOOZBYVMCGL3J/events.json","paper":"https://pith.science/paper/RFSJ3COX"},"agent_actions":{"view_html":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J","download_json":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J.json","view_paper":"https://pith.science/paper/RFSJ3COX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.05484&json=true","fetch_graph":"https://pith.science/api/pith-number/RFSJ3COXIGHRBEOOZBYVMCGL3J/graph.json","fetch_events":"https://pith.science/api/pith-number/RFSJ3COXIGHRBEOOZBYVMCGL3J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J/action/storage_attestation","attest_author":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J/action/author_attestation","sign_citation":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J/action/citation_signature","submit_replication":"https://pith.science/pith/RFSJ3COXIGHRBEOOZBYVMCGL3J/action/replication_record"}},"created_at":"2026-07-05T11:17:03.451977+00:00","updated_at":"2026-07-05T11:17:03.451977+00:00"}