{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4JN6FELGTQQ53NPADQEBEEIPMW","short_pith_number":"pith:4JN6FELG","schema_version":"1.0","canonical_sha256":"e25be291669c21ddb5e01c0812110f659e6f634894fc559ce675beadfd40f737","source":{"kind":"arxiv","id":"2602.20399","version":2},"attestation_state":"computed","paper":{"title":"GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Haixu Wu, Kaiming He, Minghao Guo, Mingsheng Long, Wojciech Matusik, Zhiyang Dou, Zongyi Li","submitted_at":"2026-02-23T22:32:08Z","abstract_excerpt":"Neural simulators promise efficient surrogates for physics simulation, but scaling them is bottlenecked by the prohibitive cost of generating high-fidelity training data. Pre-training on abundant off-the-shelf geometries offers a natural alternative, yet faces a fundamental gap: supervision on static geometry alone ignores dynamics and can lead to negative transfer on physics tasks. We present GeoPT, a unified pre-trained model for general physics simulation based on lifted geometric pre-training. The core idea is to augment geometry with synthetic dynamics, enabling dynamics-aware self-superv"},"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":"2602.20399","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-23T22:32:08Z","cross_cats_sorted":[],"title_canon_sha256":"8acc27bcb4689118069bcc62ce42f7aeb12a7e1b19f154b66be282b9cde4714a","abstract_canon_sha256":"d76973fe560c913ef4fc730bcfb8eddeb00e4c5a4a0fb2f0665a4d3e27c2f708"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:23.719404Z","signature_b64":"ySwt2CECgVl7xrwLsoK2TUkTMZaqiB6TTP9MLjiFMiKx9ZH6/7GJzhdKgWEnAg/C2Z2ZniEF6MzHu8JAD3XOCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e25be291669c21ddb5e01c0812110f659e6f634894fc559ce675beadfd40f737","last_reissued_at":"2026-05-21T01:04:23.718567Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:23.718567Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Haixu Wu, Kaiming He, Minghao Guo, Mingsheng Long, Wojciech Matusik, Zhiyang Dou, Zongyi Li","submitted_at":"2026-02-23T22:32:08Z","abstract_excerpt":"Neural simulators promise efficient surrogates for physics simulation, but scaling them is bottlenecked by the prohibitive cost of generating high-fidelity training data. Pre-training on abundant off-the-shelf geometries offers a natural alternative, yet faces a fundamental gap: supervision on static geometry alone ignores dynamics and can lead to negative transfer on physics tasks. We present GeoPT, a unified pre-trained model for general physics simulation based on lifted geometric pre-training. The core idea is to augment geometry with synthetic dynamics, enabling dynamics-aware self-superv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.20399","kind":"arxiv","version":2},"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/2602.20399/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":"2602.20399","created_at":"2026-05-21T01:04:23.718679+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.20399v2","created_at":"2026-05-21T01:04:23.718679+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.20399","created_at":"2026-05-21T01:04:23.718679+00:00"},{"alias_kind":"pith_short_12","alias_value":"4JN6FELGTQQ5","created_at":"2026-05-21T01:04:23.718679+00:00"},{"alias_kind":"pith_short_16","alias_value":"4JN6FELGTQQ53NPA","created_at":"2026-05-21T01:04:23.718679+00:00"},{"alias_kind":"pith_short_8","alias_value":"4JN6FELG","created_at":"2026-05-21T01:04:23.718679+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.16966","citing_title":"Harnessing AI for Inverse Partial Differential Equation Problems: Past, Present, and Prospects","ref_index":235,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW","json":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW.json","graph_json":"https://pith.science/api/pith-number/4JN6FELGTQQ53NPADQEBEEIPMW/graph.json","events_json":"https://pith.science/api/pith-number/4JN6FELGTQQ53NPADQEBEEIPMW/events.json","paper":"https://pith.science/paper/4JN6FELG"},"agent_actions":{"view_html":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW","download_json":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW.json","view_paper":"https://pith.science/paper/4JN6FELG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.20399&json=true","fetch_graph":"https://pith.science/api/pith-number/4JN6FELGTQQ53NPADQEBEEIPMW/graph.json","fetch_events":"https://pith.science/api/pith-number/4JN6FELGTQQ53NPADQEBEEIPMW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW/action/storage_attestation","attest_author":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW/action/author_attestation","sign_citation":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW/action/citation_signature","submit_replication":"https://pith.science/pith/4JN6FELGTQQ53NPADQEBEEIPMW/action/replication_record"}},"created_at":"2026-05-21T01:04:23.718679+00:00","updated_at":"2026-05-21T01:04:23.718679+00:00"}