{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TEZYOUS232UTBJO5QIQ67X3U7D","short_pith_number":"pith:TEZYOUS2","schema_version":"1.0","canonical_sha256":"993387525adea930a5dd8221efdf74f8f6018f7b3c523dedc2c51fcbd6f36f25","source":{"kind":"arxiv","id":"2506.10868","version":1},"attestation_state":"computed","paper":{"title":"A multi-scale loss formulation for learning a probabilistic model with proper score optimisation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"physics.ao-ph","authors_text":"Martin Leutbecher, Pedro Maciel, Simon Lang","submitted_at":"2025-06-12T16:30:18Z","abstract_excerpt":"We assess the impact of a multi-scale loss formulation for training probabilistic machine-learned weather forecasting models. The multi-scale loss is tested in AIFS-CRPS, a machine-learned weather forecasting model developed at the European Centre for Medium-Range Weather Forecasts (ECMWF). AIFS-CRPS is trained by directly optimising the almost fair continuous ranked probability score (afCRPS). The multi-scale loss better constrains small scale variability without negatively impacting forecast skill. This opens up promising directions for future work in scale-aware model training."},"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":"2506.10868","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"physics.ao-ph","submitted_at":"2025-06-12T16:30:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d2bfee3ea6275adad8d61c0a1c688a916857d4166d7aba0a103e91cf7025111e","abstract_canon_sha256":"3f70efe0b30201f32d7b111176c6c7b20d65fdb214197865f4150ff6d90d527d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:39.006109Z","signature_b64":"LTD71jbuAh3mtA5RC+hxZofKdEjWWH9EfXEaK0gwqicu/QxiFIHNRjUmLDZeJSnOOL39K4SQSsYhPJm0dSyAAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"993387525adea930a5dd8221efdf74f8f6018f7b3c523dedc2c51fcbd6f36f25","last_reissued_at":"2026-07-05T11:20:39.005761Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:39.005761Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A multi-scale loss formulation for learning a probabilistic model with proper score optimisation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"physics.ao-ph","authors_text":"Martin Leutbecher, Pedro Maciel, Simon Lang","submitted_at":"2025-06-12T16:30:18Z","abstract_excerpt":"We assess the impact of a multi-scale loss formulation for training probabilistic machine-learned weather forecasting models. The multi-scale loss is tested in AIFS-CRPS, a machine-learned weather forecasting model developed at the European Centre for Medium-Range Weather Forecasts (ECMWF). AIFS-CRPS is trained by directly optimising the almost fair continuous ranked probability score (afCRPS). The multi-scale loss better constrains small scale variability without negatively impacting forecast skill. This opens up promising directions for future work in scale-aware model training."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10868","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/2506.10868/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":"2506.10868","created_at":"2026-07-05T11:20:39.005819+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.10868v1","created_at":"2026-07-05T11:20:39.005819+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10868","created_at":"2026-07-05T11:20:39.005819+00:00"},{"alias_kind":"pith_short_12","alias_value":"TEZYOUS232UT","created_at":"2026-07-05T11:20:39.005819+00:00"},{"alias_kind":"pith_short_16","alias_value":"TEZYOUS232UTBJO5","created_at":"2026-07-05T11:20:39.005819+00:00"},{"alias_kind":"pith_short_8","alias_value":"TEZYOUS2","created_at":"2026-07-05T11:20:39.005819+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/TEZYOUS232UTBJO5QIQ67X3U7D","json":"https://pith.science/pith/TEZYOUS232UTBJO5QIQ67X3U7D.json","graph_json":"https://pith.science/api/pith-number/TEZYOUS232UTBJO5QIQ67X3U7D/graph.json","events_json":"https://pith.science/api/pith-number/TEZYOUS232UTBJO5QIQ67X3U7D/events.json","paper":"https://pith.science/paper/TEZYOUS2"},"agent_actions":{"view_html":"https://pith.science/pith/TEZYOUS232UTBJO5QIQ67X3U7D","download_json":"https://pith.science/pith/TEZYOUS232UTBJO5QIQ67X3U7D.json","view_paper":"https://pith.science/paper/TEZYOUS2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.10868&json=true","fetch_graph":"https://pith.science/api/pith-number/TEZYOUS232UTBJO5QIQ67X3U7D/graph.json","fetch_events":"https://pith.science/api/pith-number/TEZYOUS232UTBJO5QIQ67X3U7D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TEZYOUS232UTBJO5QIQ67X3U7D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TEZYOUS232UTBJO5QIQ67X3U7D/action/storage_attestation","attest_author":"https://pith.science/pith/TEZYOUS232UTBJO5QIQ67X3U7D/action/author_attestation","sign_citation":"https://pith.science/pith/TEZYOUS232UTBJO5QIQ67X3U7D/action/citation_signature","submit_replication":"https://pith.science/pith/TEZYOUS232UTBJO5QIQ67X3U7D/action/replication_record"}},"created_at":"2026-07-05T11:20:39.005819+00:00","updated_at":"2026-07-05T11:20:39.005819+00:00"}