{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YDKG43L5GDSP5NMTUZ54IXX6OZ","short_pith_number":"pith:YDKG43L5","schema_version":"1.0","canonical_sha256":"c0d46e6d7d30e4feb593a67bc45efe76466b00fa8490a431f1cf7b24b59f2175","source":{"kind":"arxiv","id":"2505.02506","version":1},"attestation_state":"computed","paper":{"title":"Exploring Design Choices for Autoregressive Deep Learning Climate Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andreas Hotho, Anna Krause, Florian Gallusser, Simon Hentschel","submitted_at":"2025-05-05T09:37:58Z","abstract_excerpt":"Deep Learning models have achieved state-of-the-art performance in medium-range weather prediction but often fail to maintain physically consistent rollouts beyond 14 days. In contrast, a few atmospheric models demonstrate stability over decades, though the key design choices enabling this remain unclear. This study quantitatively compares the long-term stability of three prominent DL-MWP architectures - FourCastNet, SFNO, and ClimaX - trained on ERA5 reanalysis data at 5.625{\\deg} resolution. We systematically assess the impact of autoregressive training steps, model capacity, and choice of p"},"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":"2505.02506","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-05T09:37:58Z","cross_cats_sorted":[],"title_canon_sha256":"f3eda850e55c91bf0e6589bd432b3909136d906cb122c4d29319523a368a3af7","abstract_canon_sha256":"643a05c944960b0811af685a95b8a2d5b50d168909e7d4f1e2576a7202a72627"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:58:38.355231Z","signature_b64":"s/Sw/espS6Ur4MH1AFhdvZLPI5kqX7eW8LbmhMBM/S6QMHzWHDRr7KSMrJeTFMvLJ6pcpnfEnoW+SqzTrMaiCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c0d46e6d7d30e4feb593a67bc45efe76466b00fa8490a431f1cf7b24b59f2175","last_reissued_at":"2026-07-05T10:58:38.354705Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:58:38.354705Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exploring Design Choices for Autoregressive Deep Learning Climate Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andreas Hotho, Anna Krause, Florian Gallusser, Simon Hentschel","submitted_at":"2025-05-05T09:37:58Z","abstract_excerpt":"Deep Learning models have achieved state-of-the-art performance in medium-range weather prediction but often fail to maintain physically consistent rollouts beyond 14 days. In contrast, a few atmospheric models demonstrate stability over decades, though the key design choices enabling this remain unclear. This study quantitatively compares the long-term stability of three prominent DL-MWP architectures - FourCastNet, SFNO, and ClimaX - trained on ERA5 reanalysis data at 5.625{\\deg} resolution. We systematically assess the impact of autoregressive training steps, model capacity, and choice of p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.02506","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/2505.02506/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":"2505.02506","created_at":"2026-07-05T10:58:38.354770+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.02506v1","created_at":"2026-07-05T10:58:38.354770+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.02506","created_at":"2026-07-05T10:58:38.354770+00:00"},{"alias_kind":"pith_short_12","alias_value":"YDKG43L5GDSP","created_at":"2026-07-05T10:58:38.354770+00:00"},{"alias_kind":"pith_short_16","alias_value":"YDKG43L5GDSP5NMT","created_at":"2026-07-05T10:58:38.354770+00:00"},{"alias_kind":"pith_short_8","alias_value":"YDKG43L5","created_at":"2026-07-05T10:58:38.354770+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/YDKG43L5GDSP5NMTUZ54IXX6OZ","json":"https://pith.science/pith/YDKG43L5GDSP5NMTUZ54IXX6OZ.json","graph_json":"https://pith.science/api/pith-number/YDKG43L5GDSP5NMTUZ54IXX6OZ/graph.json","events_json":"https://pith.science/api/pith-number/YDKG43L5GDSP5NMTUZ54IXX6OZ/events.json","paper":"https://pith.science/paper/YDKG43L5"},"agent_actions":{"view_html":"https://pith.science/pith/YDKG43L5GDSP5NMTUZ54IXX6OZ","download_json":"https://pith.science/pith/YDKG43L5GDSP5NMTUZ54IXX6OZ.json","view_paper":"https://pith.science/paper/YDKG43L5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.02506&json=true","fetch_graph":"https://pith.science/api/pith-number/YDKG43L5GDSP5NMTUZ54IXX6OZ/graph.json","fetch_events":"https://pith.science/api/pith-number/YDKG43L5GDSP5NMTUZ54IXX6OZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YDKG43L5GDSP5NMTUZ54IXX6OZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YDKG43L5GDSP5NMTUZ54IXX6OZ/action/storage_attestation","attest_author":"https://pith.science/pith/YDKG43L5GDSP5NMTUZ54IXX6OZ/action/author_attestation","sign_citation":"https://pith.science/pith/YDKG43L5GDSP5NMTUZ54IXX6OZ/action/citation_signature","submit_replication":"https://pith.science/pith/YDKG43L5GDSP5NMTUZ54IXX6OZ/action/replication_record"}},"created_at":"2026-07-05T10:58:38.354770+00:00","updated_at":"2026-07-05T10:58:38.354770+00:00"}